Conference Paper

How much can behavioral targeting help online advertising?

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Behavioral Targeting (BT) is a technique used by online advertisers to increase the effectiveness of their campaigns, and is playing an increasingly important role in the online advertising market. However, it is underexplored in academia how much BT can truly help online advertising in search engines. In this paper we provide an empirical study on the click-through log of advertisements collected from a commercial search engine. From the experiment results over a period of seven days, we draw three important conclusions: (1) Users who clicked the same ad will truly have similar behaviors on the Web; (2) Click-Through Rate (CTR) of an ad can be averagely improved as high as 670% by properly segmenting users for behavioral targeted advertising in a sponsored search; (3) Using short term user behaviors to represent users is more effective than using long term user behaviors for BT. We conducted statistical t-test which verified that all conclusions drawn in the paper are statistically significant. To the best of our knowledge, this work is the first empirical study for BT on the click-through log of real world ads.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Some of the ad selection algorithms perform ad selection based on the user data pattern [29] and the program event analysis [30]; however, the contextual and targeted advertising is treated differently as they are related to the psyche of the users. Consequently, it has been observed that the activity of users and their demographics strongly influences the ad selection, along with the user clicks of an ad [31,32]. As an example, a young female that is frequently browsing websites or using mobile apps related to the category of entertainment, would be more interested in receiving ads related to entertainment such as movies, musical instruments, etc.; consequently, it increases the click-through rates. ...
... To address data leakage issues, there are several works that propose the privacy-preserving [37,38,45] and resource-efficient mobile advertising systems [24,43]. The primary focus of mobile ads characterisation is on measuring the efficiency of targeted advertising and to evaluate improved performance of targeted advertising for click-through rates [31]. However, we note that there are limited insights about evaluating the effectiveness of targeting mobile advertising that will ultimately determine the magni-tude of various issues, e.g. ...
... There are a number of reasons why the existing inbrowser [6,31,37,38,[46][47][48][49][50][51] ads characterisation approaches on targeted advertisements cannot be directly applied to the evaluation of in-app targeted ads: First, there may be various factors for in-app ads targeting that go beyond what is considered for in-browser ads, e.g. the context of mobile apps installed on the user device, their utilisation behaviour (e.g. heavy gamers may receive specific ads). ...
Article
Full-text available
Targeted advertising has transformed the marketing landscape for a wide variety of businesses, by creating new opportunities for advertisers to reach prospective customers by delivering personalised ads, using an infrastructure of a number of intermediary entities and technologies. The advertising and analytics companies collect, aggregate, process, and trade a vast amount of users’ personal data, which has prompted serious privacy concerns among both individuals and organisations. This article presents a comprehensive survey of the privacy risks and proposed solutions for targeted advertising in a mobile environment. We outline details of the information flow between the advertising platform and ad/analytics networks, the profiling process, the measurement analysis of targeted advertising based on user’s interests and profiling context, and the ads delivery process, for both in-app and in-browser targeted ads; we also include an overview of data sharing and tracking technologies. We discuss challenges in preserving the mobile user’s privacy that include threats related to private information extraction and exchange among various advertising entities, privacy threats from third-party tracking, re-identification of private information and associated privacy risks. Subsequently, we present various techniques for preserving user privacy and a comprehensive analysis of the proposals based on such techniques; we compare the proposals based on the underlying architectures, privacy mechanisms, and deployment scenarios. Finally, we discuss the potential research challenges and open research issues.
... A lot of previous research discussed how to increase the conversion rate. Yan et al. (2009) suggested the behavioral targeting tactic could make the campaigns more effective. They discovered that by correctly segmenting consumers for behaviorally targeted advertising, ads CTR (Click-Through rate) could be boosted on average by as much as 670% [4]. ...
... Yan et al. (2009) suggested the behavioral targeting tactic could make the campaigns more effective. They discovered that by correctly segmenting consumers for behaviorally targeted advertising, ads CTR (Click-Through rate) could be boosted on average by as much as 670% [4]. On top of this, Beales (2010) also indicated that behaviorally targeted advertising is more effective than traditional runof-network advertising since it provides consumers with more useful adverts [5]. ...
Article
Full-text available
The conversion rate is important measurable data as a metric of the effectiveness of advertising. The higher the conversion rate is, the more people who saw the advertising would take action. Many factors could affect the advertising conversion rate. In this paper, we aimed to find whether the format (video and poster) of advertising presentations impacts the advertising conversion rates among college students. The researcher designed an online survey that was distributed to college students, and respondents were asked to evaluate eight products’ ads with both video and poster types. In addition we investigated five potential factors, including the level of impact social media has on advertising, gender, preferences of activities, time spent on different advertising channels, and time from viewing the ads before buying products, that would influence people’s conversion rates. The results suggested that the conversion rates between video and poster are different ( p<0.05). However, all five elements had no correlations with the conversion rate.
... The direct value that consumers derive from behavioral advertising, however, has been more often posited (Beales, 2010) than empirically demonstrated. 2 Behaviorally targeted ads tend to receive higher click-through rates than non targeted ones (Bleier & Eisenbeiss, 2015;Yan et al., 2009), suggesting that the former can reduce consumer search costs. Other than through such cost reduction, however, little is known about the manner and extent to which targeted ads affect consumers' welfare. ...
... Most studies on targeted advertising have taken a narrow view with respect to its effectiveness. Such studies evaluate ad effectiveness largely in terms of click-through rates (Bleier, & Eisenbeiss, 2015;Yan et al., 2009), purchase intentions (Van Doorn & Hoekstra, 2013; Bart et al., 2012), or purchase probability (Manchanda et al., 2006;Lewis & Reily, 2009), which are higher for targeted relative to untargeted ads. Yet these effects, too, are nuanced. ...
Conference Paper
Full-text available
The value that consumers derive from behavioral advertising has been more often posited than empirically demonstrated. The majority of empirical work on behavioral advertising has focused on estimating the effectiveness of behaviorally targeted ads, measured in terms of click or conversion rates. We present the results of two online within-subject experiments (Study 1 and Study 2) that, instead, employ a counterfactual approach, designed to assess comparatively some of the consumer welfare implications of behaviorally targeted advertising. Participants are presented with alternative product offers: products advertised in ads displayed to them on websites that commonly show behaviorally targeted ads (ad condition); competing products from the organic results of online searches (search condition); and random products (random condition). The alternatives are compared along a variety of metrics, including objective measures (such as product price and vendor quality) and participants’ self-reports (such as purchase intention and perceived product relevance). In Study 1 (n = 489) we find, first, that both ads and organic search results within our sample of participants are dominated by a minority of vendors; however, ads are more likely to present participants with less popular (and therefore lesser known) vendors. Second, we find that purchase intentions are higher in the ad and the search conditions than in the random condition; the effect is driven by higher product relevance in the ad and search conditions; however, in absolute terms, product relevance is low, even in the ad condition. Third, we find that ads are more likely to be associated with lower quality vendors, and higher prices (for identical products), compared to competing alternatives found in search results. Study 2 (n = 493) replicates Study 1 results. In addition, Study 2 finds that higher purchase intentions and higher relevance in the ad condition are driven by participants having previously searched for the advertised product. Furthermore, we use a latent utility model to estimate differences in consumer surplus (a commonly used measure of consumer welfare) across conditions. In our sample of participants, the random condition is associated with the lowest surplus. After accounting for differences in vendor quality, the search condition is associated with slightly higher surplus relative to the ad condition.
... Nevertheless, the ability to target advertising to individual consumers is one of the crucial factors responsible for the generation of large revenues in the online advertising market [19,32,35,36,53,110]. Targeting refers to advertisers' ability to match ads to Internet users in the attempt to meet their preferences and interests. ...
... Across policy and academic circles, contrasting propositions have been offered regarding the effects of online advertising (including targeted advertising) on the welfare of different stakeholders (consumers, online publishers, advertising vendors, and data companies). One the one hand, some studies show a positive impact of targeting on advertising campaigns' effectiveness, such as click-through and conversion rates, website visits, and sales [19,32,35,36,53,110]. On the other hand, other researchers (and even some advertisers [101]) argue that the effect of targeted ads on consumers' likelihood to purchase may be overestimated due to "activity bias" [67], and methodological issues [32,66,85] such as large confidence intervals and (sometimes) absence of comparisons with a randomly selected control group. ...
Conference Paper
Full-text available
Ad-blocking applications have become increasingly popular among Internet users. Ad-blockers offer various privacy-and security-enhancing features: they can reduce personal data collection and exposure to malicious advertising, help safeguard users' decision-making autonomy, reduce users' costs (by increasing the speed of page loading), and improve the browsing experience (by reducing visual clutter). On the other hand, the online advertising industry has claimed that ads increase consumers' economic welfare by helping them find better, cheaper deals faster. If so, using ad-blockers would deprive consumers of these benefits. However, little is known about the actual economic impact of ad-blockers. We designed a lab experiment (N=212) with real economic incentives to understand the impact of ad-blockers on con-sumers' product searching and purchasing behavior, and the resulting consumer outcomes. We focus on the effects of blocking contextual ads (ads targeted to individual, potentially sensitive, contexts, such as search queries in a search engine or the content of web pages) on how participants searched for and purchased various products online, and the resulting consumer welfare. We find that blocking contextual ads did not have a statistically significant effect on the prices of products participants chose to purchase, the time they spent searching for them, or how satisfied they were with the chosen products, prices, and perceived quality. Hence we do not reject the null hypothesis that consumer behavior and outcomes stay constant when such ads are blocked or shown. We conclude that the use of ad-blockers does not seem to compromise consumer economic welfare (along the metrics captured in the experiment) in exchange for privacy and security benefits. We discuss the implications of this work in terms of end-users' privacy, the study's limitations, and future work to extend these results.
... As a developing phenomenon, there are neither strong definitions of behavioural targeting nor an evident accumulation of empirical findings on the effect of Ads intrusiveness on the perception of behavioural targeting and advertising outcomes such as click-through intentions (Boerman et al., 2017), click-through ratio (Yan et al., 2009), sales conversion (Farahat & Bailey, 2012), purchase intention and actual purchases (Boerman et al., 2017;Fachryto & Achyar, 2018) and revenue (Beales, 2011;Breznitz & Palermo, 2013). Although behavioural targeting is expected to attract users' interest to more personalized and relevant Ad content, not so much is known about the effect of such behavioural targeting characteristics on the perception of behavioural targeting and advertising outcomes. ...
... Kumar and Patel (2014), for example, found that targeted Ads that kept in view the interest of the internet users and their browsing history were more likely to get positive responses (click through), consolidating the claim of Kaspar et al. (2019) who reported that attention for personally relevant advertisement can be strong. Yan et al. (2009) also provided an empirical investigation on the clickthrough log of advertisements retrieved from a commercial search engine which showed that click-through rates can be averagely increased by 670% by properly segmenting web users based on their short-term online behaviour for behavioural targeting. So, we hypothesize that: ...
Article
Full-text available
This study assessed the effect of adverts (Ads) based on users’ interest, Ads intrusiveness on the perception of behavioural targeting and click-through intention. The study further examined how the perception of behavioural targeting mediates the interaction between Ads based on users’ interest, Ads intrusiveness and click-through intentions. The study was a descriptive survey, and the approach was deductive. The study collected primary data from a sample of 376 internet users in Nigeria using a cloud-based questionnaire. The data obtained were analysed using descriptive and inferential statistical tools. Partial least square structural equation modelling (PLS-SEM) was used to test the hypotheses. Findings showed that Ads based on users’ interest and Ads intrusiveness predict click-through intentions and the perception of behavioural targeting. Perception of behavioural targeting mediated the interaction between Ads based on users’ interests and click-through intentions. Overall, the study concluded that Ads based on users’ interest and Ads intrusiveness influences the perception of behavioural targeting, which affects click�through intention. Hence the study recommended that digital entrepreneurs carefully design, manage and control the Ads contents they direct to customers. Digital entrepreneurs should exercise caution in implementing their behavioural targeting strategy so as not to breach an acceptable threshold.
... To solve the incompleteness problem, we resort to other signals besides pure implicit feedback from users. We consider modelfree methods like behavioral retargeting (BR) [43] and item-based collaborative filtering (ItemCF) [33], which are widely used in ecommerce recommender systems. These methods are different from future behavior prediction in neural sequential recommendation models: BR recommends a user with items that are in the historical behaviors and ItemCF recommends items similar to the historical behaviors where the similarity between items is measured globally. ...
... We investigate its influence on neural sequential recommendation models by comparing the top-set recommended by models trained according to future behaviors to that recommended by model-free methods which recommend items in a different way from future behavior prediction. We consider behavioral retargeting (BR) [43] and item-based collaborative filtering (ItemCF) [33] since both of them prevail in practice, though our method proposed in Section 5 does not limit to BR and ItemCF. ...
Preprint
Learning the user-item relevance hidden in implicit feedback data plays an important role in modern recommender systems. Neural sequential recommendation models, which formulates learning the user-item relevance as a sequential classification problem to distinguish items in future behaviors from others based on the user's historical behaviors, have attracted a lot of interest in both industry and academic due to their substantial practical value. Though achieving many practical successes, we argue that the intrinsic {\bf incompleteness} and {\bf inaccuracy} of user behaviors in implicit feedback data is ignored and conduct preliminary experiments for supporting our claims. Motivated by the observation that model-free methods like behavioral retargeting (BR) and item-based collaborative filtering (ItemCF) hit different parts of the user-item relevance compared to neural sequential recommendation models, we propose a novel model-agnostic training approach called WSLRec, which adopts a three-stage framework: pre-training, top-$k$ mining, and fine-tuning. WSLRec resolves the incompleteness problem by pre-training models on extra weak supervisions from model-free methods like BR and ItemCF, while resolves the inaccuracy problem by leveraging the top-$k$ mining to screen out reliable user-item relevance from weak supervisions for fine-tuning. Experiments on two benchmark datasets and online A/B tests verify the rationality of our claims and demonstrate the effectiveness of WSLRec.
... This is because users' activities on the social network (e.g., browsing history and updated status) reveal their personal characteristics and preferences. By exploiting such valuable information, the SNP can present the advertisements effectively to those who are more likely to be interested in the related products [9], [10]. For example, showing luxury product advertisements to wealthy users would be more effective than to average users. ...
... The proof of Proposition 5 is in Appendix E. If the probability P H is higher than V (δx * L ) /V (δx * H ), the optimal advertisement price is the valuation V (δx * H (δ)) associated with high-type users, in which case the SNP can 9 The main reason of considering discrete values is to derive the closed-form solution of the SNP's optimal advertisement price and privacy policy. ...
Article
In an online social network, users exhibit personal information to enjoy social interaction. The social network provider (SNP) exploits users' information for revenue generation through targeted advertisement, in which the SNP presents advertisements to proper users effectively. Therefore, an advertiser is more willing to pay for targeted advertisement to promote his product. However, the over-exploitation of users' information would invade users' privacy, which would negatively impact users' social activeness. Motivated by this, we study the privacy policy (policies) of the SNP(s) with targeted advertisement, in both monopoly and duopoly markets. We characterize the privacy policy in terms of the fraction of users' information that the provider should exploit, and formulate the interactions among users, advertiser, and SNP(s) as a three-stage Stackelberg game. By leveraging the model's supermodularity property, we prove the threshold structure of users' equilibrium information levels. We discover the overall information that can be exploited by an SNP is non-monotonic in the exploitation fraction. Monopoly (one SNP) study shows our proposed optimal privacy policy helps the SNP earn even more advertisement revenue than full exploitation policy does. The situation of the duopoly market is much more complicated. In that case, if the service quality gap between the two SNPs is large, the stronger SNP will choose a conservative privacy protection policy that drives the other SNP out of the market. However, if the service quality gap is small and the advertisement revenue is promising, the stronger SNP would choose an aggressive policy to exploit the advertisement revenue and both SNPs will have positive market shares.
... This framework forms the foundation, guiding the study's research steps, and provides the basis for interpreting the research results. Future research efforts would focus on empirically testing these hypotheses to either corroborate or adjust the proposed framework [54,55]. ...
Article
Full-text available
The accelerating advancement of Artificial Intelligence (AI) technologies has brought forth a myriad of opportunities in many sectors, with digital marketing being a particularly prominent area of application. This study titled "Personalization and Profits: The Impact of AI on Targeted Digital Marketing" elucidates the profound influence of AI on personalized, targeted marketing strategies and, consequently, its influence on corporate profit growth. The paper leverages empirical data gathered from multiple prominent business organizations and conducts an in-depth analysis to demonstrate how AI has transformed marketing from a broad-scoped field to a focused, person-centric medium. It extends its investigation on how such targeted, personalized digital marketing efforts have increased customer engagement, boosted conversion rates, and notably impacted profitability. The dynamic AI algorithms' ability to mine, process, and draw actionable insights from vast troves of consumer data, enables unprecedented levels of marketing personalization. The study shows how such personalization strategies galvanized by AI have significantly enhanced marketing efficiency, and therefore have led to higher profit margins. The study underscores the importance of integrating AI in any digital marketing strategy, not merely to stay relevant in the ever-evolving market but also for ensuring continued profitability. It ultimately provides a comprehensive view of how AI's broad-ranging, transformative effect on targeted digital marketing has a critical impact on business profits in today's hyper-competitive market landscape.
... By understanding the user behaviour, designers and marketers can improve the advertisement strategy, increase the click rates, and enhance the effectiveness of interactive media. In [8] an empirical study over the ads click-through log (collected from a commercial search engine) is presented. After analysing the results, the following set of conclusions can be highlighted: (1). ...
Article
Full-text available
In the past few years, the online video streaming market has witnessed rapid growth and has become the most important form of entertainment. Motivated by the huge business opportunities, the advertisement insertion mechanisms have become a hot topic of research and represent the most important component of an online delivery ecosystem. In this paper, we introduce SemanticAd , a multimodal ad insertion framework designed from the viewers’ perspective in terms of the quality of experience and degree of intrusiveness. The core of the proposed approach involves a novel temporal segmentation algorithm that extracts story units with a frame level precision. To the best of our knowledge, the proposed solution is the most robust and accurate solution dedicated to TV news videos. In addition, by taking into consideration ad temporal distribution and semantic information, the framework proposes commercials that are contextually relevant with respect to video content. The quantitative and qualitative experimental results conducted on a challenging set of 50 multimedia documents validate the SemanticAd methodology, returning a F1-score superior to 92%. Moreover, when compared to other state-of-the-art methods, our system demonstrates its superiority with gains in performance ranging in the [4.19%, 10.22%] interval.
... Due to the high expense of such initiatives, management is forced to limit inducement to customers who are most likely to exhibit the behavior of interest. As another example, companies are interested in identifying users who would click on an advertisement if and only if it is highlighted in online advertising (Yan et al. 2009;Bottou et al. 2013;Li et al. 2014;Sun et al. 2015). The challenge in identifying these users stems from the fact that the desired response pattern is not observed directly but rather is defined counterfactually in terms of what the individual would do under hypothetical unrealized conditions. ...
Article
The unit selection problem aims to identify a set of individuals who are most likely to exhibit a desired mode of behavior or to evaluate the percentage of such individuals in a given population, for example, selecting individuals who would respond one way if encouraged and a different way if not encouraged. Using a combination of experimental and observational data, Li and Pearl solved the binary unit selection problem (binary treatment and effect) by deriving tight bounds on the "benefit function," which is the payoff/cost associated with selecting an individual with given characteristics. This paper extends the benefit function to the general form such that the treatment and effect are not restricted to binary. We then propose an algorithm to test the identifiability of the nonbinary benefit function and an algorithm to compute the bounds of the nonbinary benefit function using experimental and observational data.
... More specifically, these studies find positive effects of targeting on a variety of outcome measures, including the following: click-through rates (Yan, Liu, Wang, Zhang, Jiang, & Chen, 2009), which depend on timing and placement factors (Bleier & Eisenbeiss, 2015); consumers' purchase intentions (van Doorn & Hoekstra, 2013); purchase probabilities (Manchanda, Dubé, Goh, & Chintagunta, 2006); consumers' progression through the purchase funnel (Hoban & Bucklin, 2015); ad prices and ad revenue (Beales, 2010), ad revenue per impression (Ada, Nabout, & Feit, 2022); and advertising profitability (Lewis & Reiley, 2014). These varying outcomes also explain differences in the value of users for publishers. ...
... Some scholars claim that consumers perceive targeted ads as more relevant, motivating, and appealing (Tucker, 2014). This view is empirically supported by studies revealing positive attitudes towards personalised advertising (Maslowska et al., 2011), higher click-through rates online (Yan et al., 2009) and increased purchase decision-making (Goldfarb & Tucker, 2012). In practice, advertising practitioners have found that addressable TV can considerably improve a campaign's impact by lowering channel switching, increasing resonance of brand messaging, improving recall, and enhancing purchase intent (Sky, 2019). ...
Article
Full-text available
As TV consumption evolves from traditional linear programming to more on-demand viewing, advertising is also changing, seeking to tailor content to best match the interests of viewers. Addressable advertising is an interactive form of advertising that combines online data personalization with on-demand TV content with the aim of addressing individual viewers and improving advertising outcomes. This study investigated whether audience engagement with advertising (indexed by self-report liking, attention, and memory for an advertisement) was affected by addressability and the screen size on which the content was viewed. Using a limited capacity model of information processing and the elaboration likelihood model as its theoretical bases as well as a physiological measure of attention, we found that people both prefer and remember addressable advertisements more than those that are not relevant to them. In addition, viewing advertisements on large screens improved attention and retention for the content relative to smaller screens.
... The performance-based models, such as pay-perclick, rely on a click (or a sale, application, or lead). As clickthrough rates for online advertising declined due to competition, websites slowly turned to targeted advertising to increase their ad effectiveness and maintain their revenues (Boerman et al., 2017;Farahat & Bailey, 2012;Yan et al., 2009). Targeted advertising can be done through the content of a webpage, which is one of the simplest techniques, or through a user's online behavior, one of the advanced techniques (Ur et al., 2012). ...
Article
Full-text available
Advertising has been the most common strategy for content monetization. However, behavioral data collection for personalized advertisements has raised privacy concerns. This study explores users’ perceptions of browser-based cryptocurrency mining (BCM) as an alternative content monetization method to online advertising. The mining process is an essential part of blockchain technology, but it is a resource-intensive process. By utilizing BCM, content providers can mine cryptocurrency on the computers of website visitors and provide free or discounted services for the cost of the mining process. Considering that BCM revenue can be comparable to online advertisement while preserving user privacy at a negligible cost, we explored underlying factors affecting the adoption of BCM. We conducted a qualitative exploratory study using linguistic analysis and topic modeling of data collected from an online focus group. The results revealed that users raised security concerns and resisted BCM adoption due to its negative image and potential risk barriers. Our results further showed that online advertising benefits from the status quo bias. That is, although users raised both privacy and security concerns, they still considered online advertising. We extend the literature by analyzing controversial technology in the light of incumbent technology and provide directions for implementing BCM as a successful revenue stream.
... For example, if the target is the young producer, they should post on the platform of Bilibili instead of WeChat according to the expected results. "Click-Through Rate (CTR) of an ad can be averagely improved as high as 670% by properly segmenting users for behaviorally targeted advertising in a sponsored search" [10]. Therefore, choosing and arranging the targeted advertisement and consumer is quite important to increase sales ...
Article
Full-text available
Since the increasing importance of social media online advertisement, the area has existed a large amount of research. The majority of the paper focuses on the effectiveness and content of online advertisements. However, the relationship between social media platforms and the effectiveness of online advertisement seems to be less concentrated in this area. This paper seeks to solve this issue in the online advertisement by providing a methodology and a way to measure and analyze the data. The methodology used in this paper is a survey. This paper focuses on two particular platforms, Bilibili and WeChat, and the comparison is made in three specific kinds of products, including electronic devices, food, and clothes. Assuming Bilibili and WeChat target different groups of people in different areas, this paper further provides suggestions to the enterprise. This paper also includes some suggestions for increasing the social media platform performance and boosting the consumers’ purchase intention.
... Social media users make many types of personal information publicly available [2]. Firms use that information to learn more about potential customers and target advertisements accordingly [32], sometimes influencing end-users [20] without their explicit consent or awareness -a form of hidden "digital market manipulation" [7]. Leveraging individuals' innate attraction to selfmorphs to promote products is an example of a targeted marketing strategy [10] that may influence end-users' actions while operating outside their awareness, raising potential yet significant privacy concerns. ...
Conference Paper
Full-text available
Self-images are among the most prevalent forms of content shared on social media streams. Face-morphs are images digitally created by combining facial pictures of different individuals. In the case of self-morphs, a person's own picture is combined with that of another individual. Prior research has shown that even when individuals do not recognize themselves in self-morphs, they tend to trust self-morphed faces more, and judge them more favorably. Thus, self-morphs may be used online as covert forms of targeted marketing-for instance, using consumers' pictures from social media streams to create self-morphs, and inserting the resulting self-morphs in promotional campaigns targeted at those consumers. The usage of this type of personal data for highly targeted influence without individuals' awareness, and the type of opaque effect such artifacts may have on individuals' attitudes and behaviors, raise potential issues of consumer privacy and autonomy. However, no research to date has examined the feasibility of using self-morphs for such applications. Research on self-morphs has focused on artificial laboratory settings, raising questions regarding the practical, in-the-wild applicability of reported self-morph effects. In three experiments, we examine whether self-morphs could affect individuals' attitudes or even promote products/services, using a combination of experimental designs and dependent variables. Across the experiments, we test both designs and variables that had been used in previous research in this area and new ones that had not. Questioning prior research, however, we find no evidence that end-users react more positively to self-morphs than control-morphs composed of unfamiliar facial pictures in either attitudes or actual behaviors.
... For example, in regard to the scenario "Marketing companies can use fitness tracker data in order to send you specific advertisements regarding running shoes", 68.2% of the users reported this as "Very likely to happen" and 18.2% as "Likely to Happen", while none of the respondents responded with "Very unlikely to happen" or "Unlikely to happen". This is quite predicted as online targeted advertising has shown great market potential (Yan et al. 2009) and is widely used today. In another case, the scenario "A murder can be solved by using the victim's fitness tracker data, such as heart rate data" has been acknowledged as "Very likely to happen" by 54.5% and as "Likely to Happen" by 27.3% by the participants (none of the participants responded with "Very unlikely to happen" or "Unlikely to happen"). ...
Article
Full-text available
In the IoT era, sensitive and non-sensitive data are recorded and transmitted to multiple service providers and IoT platforms, aiming to improve the quality of our lives through the provision of high-quality services. However, in some cases these data may become available to interested third parties, who can analyse them with the intention to derive further knowledge and generate new insights about the users, that they can ultimately use for their own benefit. This predicament raises a crucial issue regarding the privacy of the users and their awareness on how their personal data are shared and potentially used. The immense increase in fitness trackers use has further increased the amount of user data generated, processed and possibly shared or sold to third parties, enabling the extraction of further insights about the users. In this work, we investigate if the analysis and exploitation of the data collected by fitness trackers can lead to the extraction of inferences about the owners routines, health status or other sensitive information. Based on the results, we utilise the PrivacyEnhAction privacy tool, a web application we implemented in a previous work through which the users can analyse data collected from their IoT devices, to educate the users about the possible risks and to enable them to set their user privacy preferences on their fitness trackers accordingly, contributing to the personalisation of the provided services, in respect of their personal data.
... For example, decision-makers in customer relationship management want to identify customers who would buy purchases if there is an enticement and would not otherwise [1,6,7,22]. For another example, large companies are interested in customers who would visit their website if there the website is prompted by an online advertisement and would not otherwise [2,13,17,20,23]. The challenge of identifying such customers stems from the fact that the desired behaviors are defined counterfactually and would occur under hypothetical space. ...
Preprint
Full-text available
The unit selection problem defined by Li and Pearl identifies individuals who have desired counterfactual behavior patterns, for example, individuals who would respond positively if encouraged and would not otherwise. Li and Pearl showed by example that their unit selection model is beyond the A/B test heuristics. In this paper, we reveal the essence of the A/B test heuristics, which are exceptional cases of the benefit function defined by Li and Pearl. Furthermore, We provided more simulated use cases of Li-Pearl's unit selection model to help decision-makers apply their model correctly, explaining that A/B test heuristics are generally problematic.
... The incentive should be sent very carefully because the behavior of the customers is influenced by any incentives that the companies sent before. For another example, in online advertisement, companies are interested in identifying users who would view an advertisement if and only if the advertisement is prompted [2,12,16,18,20]. The challenge in identifying these individuals stems from the fact that the desired response pattern is not observed directly but rather is defined counterfactually in terms of what the individual would do under hypothetical unrealized conditions. ...
Preprint
Full-text available
The unit selection problem is to identify a group of individuals who are most likely to exhibit a desired mode of behavior, for example, selecting individuals who would respond one way if incentivized and a different way if not. The unit selection problem consists of evaluation and search subproblems. Li and Pearl defined the "benefit function" to evaluate the average payoff of selecting a certain individual with given characteristics. The search subproblem is then to design an algorithm to identify the characteristics that maximize the above benefit function. The hardness of the search subproblem arises due to the large number of characteristics available for each individual and the sparsity of the data available in each cell of characteristics. In this paper, we present a machine learning framework that uses the bounds of the benefit function that are estimable from the finite population data to learn the bounds of the benefit function for each cell of characteristics. Therefore, we could easily obtain the characteristics that maximize the benefit function.
... Many online retailers use retargeting, because, according to Criteo, companies that specialize in determining the target audience based on user actions, accumulating information about them, 90% of visitors leave the site without making a purchase. Taking into account that 70% of visitors are directed to e-Commerce sites by a link to paid advertising, we can say that thanks to this method, the conversion rate or index of the number of site visitors increases dramatically [8]. But targeted advertising has its drawbacks: • A small level of motivation to go to the link •. ...
Chapter
This article discusses targeted advertising. Targeted advertising in our minds has led to the fact that it is between contextual and media advertising. On the one hand, targeted advertising is based on the principles of bid management, bid management, the need to regulate targeting, and SPS models. On the other hand, targeted advertising provides for a wide coverage, a wide range of visions, and an increase in search demand. Targeted advertising in our minds has led to the fact that it is between contextual and display advertising. This article discusses the advantages, disadvantages, and advantages of targeted advertising placement systems, how to place them, what the scale should be, and what the target audience is intended for placement.KeywordsTargeted advertisingOnline advertisingSocial networkOnline advertisingSystem analysisPlacement systemDigital advertising
... In the RTB world, an Ad Platform is responsible for selecting the type of users that suit the advertiser's needs better. To achieve a better match between an advertiser and a publisher, data-mining algorithms are used that allow building a wide user's profile [12]- [14]. At some point a disbalance emerged between Ad Platforms: some had more advertisers but lacked publishers, while others had more publishers than there were advertisements available. ...
Article
Full-text available
In the modern online world, users are often asked for a permission to track their actions as a permission to “allow cookies”. The gathered information could be very valuable for a potential advertiser. However, online tracking is not only a benefit for a user but also a threat to the user’s privacy. This information combined with a targeted advertisement on a mass scale has potential to alter behaviour of large groups. This study summarises previous academic work on online user tracking and anti-tracking measures. As a result, it describes the current mechanisms used to track a user, as well as some methods that can be applied to reduce tracking. The study concludes that government legislation and open dialog between Internet users and advertisers might be the only way to ensure online privacy.
... targeting enhances advertising response rates (Yan, Liu, Wang, et al., 2009). An additional benefit also likely may arise in terms of advertisement attribution, or the process of tracking the effects of advertisements and advertising campaigns. ...
... Due to the high expense of such initiatives, management is forced to limit inducement to customers who are most likely to exhibit the behavior of interest. As another example, companies are interested in identifying users who would click on an advertisement if and only if it is highlighted in online advertising (Yan et al. 2009;Bottou et al. 2013;Li et al. 2014;Sun et al. 2015). The challenge in identifying these users stems from the fact that the desired response pattern is not observed directly but rather is defined counterfactually in terms of what the individual would do under hypothetical unrealized conditions. ...
Article
The unit selection problem aims to identify a set of individuals who are most likely to exhibit a desired mode of behavior, for example, selecting individuals who would respond one way if encouraged and a different way if not encouraged. Using a combination of experimental and observational data, Li and Pearl derived tight bounds on the "benefit function" - the payoff/cost associated with selecting an individual with given characteristics. This paper shows that these bounds can be narrowed significantly (enough to change decisions) when structural information is available in the form of a causal model. We address the problem of estimating the benefit function using observational and experimental data when specific graphical criteria are assumed to hold.
... Many online retailers use retargeting, because, according to Criteo, companies that specialize in determining the target audience based on user actions, accumulating information about them, 90% of visitors leave the site without making a purchase. Taking into account that 70% of visitors are directed to e-Commerce sites by a link to paid advertising, we can say that thanks to this method, the conversion rate or index of the number of site visitors increases dramatically [8]. But targeted advertising has its drawbacks: • A small level of motivation to go to the link •. ...
Chapter
Full-text available
In the era of globalization of the world economy, many sectors of the economy are being transformed, and agriculture is the engine of economic development in developing countries. Countries with developing economies are adopting the experience of developed countries and are beginning to apply innovative, digital technologies in agriculture, which is a key tool in increasing the productivity and efficiency of the industry. Of course, the digitalization of agriculture occupies one of the leading positions in this issue, because the economic stability of the state largely depends on the degree of development of the agricultural sector in the country. Agriculture in the world is turning from a traditional to a high-tech industry, which can create new markets for innovative solutions and developments. The results of the study: (1) the analysis of digital technologies in agriculture is presented; (2) methods and techniques of farming through digital technologies are disclosed; (3) the concepts reflecting various forms of digitalization in the agricultural sector are considered; (4) the economic efficiency of grain production in the regions of Kazakhstan is determined; (5) various applications of digital technologies and innovations in agriculture and farming are shown.KeywordsDigital technologiesInnovationsAgricultural sectorDigitalizationSustainable agricultureStrategic developmentEconomicsInnovative developmentsJEL CodesO10O20O30P33P47
... The advertising technology industry uses personalization, a form of data extraction, to collect user behavioral data in order to increase monetization [81]. A majority of ad categories are behaviorally targeted [46] which can increase click-through rates up to 670% [78]. Furthermore, health ads rely on re-marketing, meaning that ads are delivered frequently to users expressing specific interest in health-related products [46]. ...
Preprint
Full-text available
Targeted advertising can harm vulnerable groups when it targets individuals' personal and psychological vulnerabilities. We focus on how targeted weight-loss advertisements harm people with histories of disordered eating. We identify three features of targeted advertising that cause harm: the persistence of personal data that can expose vulnerabilities, over-simplifying algorithmic relevancy models, and design patterns encouraging engagement that can facilitate unhealthy behavior. Through a series of semi-structured interviews with individuals with histories of unhealthy body stigma, dieting, and disordered eating, we found that targeted weight-loss ads reinforced low self-esteem and deepened pre-existing anxieties around food and exercise. At the same time, we observed that targeted individuals demonstrated agency and resistance against distressing ads. Drawing on scholarship in postcolonial environmental studies, we use the concept of slow violence to articulate how online targeted advertising inflicts harms that may not be immediately identifiable. CAUTION: This paper includes media that could be triggering, particularly to people with an eating disorder. Please use caution when reading, printing, or disseminating this paper.
... Modern search engines store user queries in the query log. The query log is frequently used in the advanced information retrieval (IR) algorithms for refining the retrieved information with the help of personalized information retrieval (IR) [8]- [10]. Although search engines use query log for personalized IR, query log may cause privacy concerns, revealing a lot of private information about individuals if used against them [6], [11], [12]. ...
Article
Full-text available
Search engines store users’ queries in a query log for performing personalized information retrieval. However, query logs cause privacy concerns and reveal a lot of information about individuals if used against them. Private web search (PWS) provides a privacy-preserving information retrieval (IR) facility which allows users to retrieve information from an IR system without revealing true search queries. Current PWS techniques that are explored in the domain of web search are query obfuscation-based private web search (OB-PWS). These techniques achieve web privacy by injecting cover queries into the user profiles. However, existing OB-PWS techniques submit true queries along with cover queries and achieve query obfuscation in an isolated manner without considering the similarity between consecutive queries. In this article, we propose a proxy-terms based query obfuscation technique that allows users to retrieve information from an IR system through proxy queries without submitting true queries. IR system automatically generates cover queries and true queries from the proxy queries and cannot differentiate whether the user is trying to retrieve information for the cover queries or true query. We analyzed the effectiveness of the proposed technique on test queries, and develop a similarity metric for testing the accuracy of the proposed technique. Promising results of experiments confirm the effectiveness of the proposed technique.
... Representing and predicting user behavior has long been a topic of interest in Machine Learning. It is a critical issue for many applications, such as Churn Detection (Berger and Kompan 2019;Pudipeddi et al. 2014;Kwon et al. 2019), Advertising (Tu and Lu 2010;Yan et al. 2009), Social Media (Hallac et al. 2019;Wang et al. 2019) and more. ...
Thesis
Full-text available
The last decade has witnessed the impressive expansion of Deep Learning (DL) methods, both in academic research and the private sector. This success can be explained by the ability DL to model ever more complex entities. In particular, Representation Learning methods focus on building latent representations from heterogeneous data that are versatile and re-usable, namely in Natural Language Processing (NLP). In parallel, the ever-growing number of systems relying on user data brings its own lot of challenges. This work proposes methods to leverage the representation power of NLP in order to learn rich and versatile user representations.Firstly, we detail the works and domains associated with this thesis. We study Recommendation. We then go over recent NLP advances and how they can be applied to leverage user-generated texts, before detailing Generative models.Secondly, we present a Recommender System (RS) that is based on the combination of a traditional Matrix Factorization (MF) representation method and a sentiment analysis model. The association of those modules forms a dual model that is trained on user reviews for rating prediction. Experiments show that, on top of improving performances, the model allows us to better understand what the user is really interested in in a given item, as well as to provide explanations to the suggestions made.Finally, we introduce a new task-centered on UR: Professional Profile Learning. We thus propose an NLP-based framework, to learn and evaluate professional profiles on different tasks, including next job generation.
... The democratisation of technology is making it imperative for media organisations to focus not merely on the demographic information of consumers, for segmentation etc., but also their media consumption habits and use behavioural targeting strategies to provide consumers with information about entertainment content that would interest them. Behavioral targeting (Yan et al., 2009) refers to gathering data about a consumer's online browsing and shopping behaviour and then sending him product related information that is relevant to him, to increase chances of purchase. In the context of the M&E industry, the viewer's past media consumption behaviour can give a service provider adequate information about his preferences. ...
Article
Full-text available
Technology is the mast that keeps the flag of the entertainment industry flying high. As technology evolves and the mode and quality of entertainment changes vigorously, India represents one of the Top 5 Entertainment and Media markets across the world. The growing usage of 3G, 4G and portable devices, rising advertising revenues, growing consumer demand has resulted in the growth of the Indian media and entertainment industry at a CAGR of 10.9 percent from FY 17-18. Constant support from TRAI (Telecom Regulatory authority of India) and the Govt. of India has pushed the Indian Media and Entertainment(M&E) Industry on a progressive growth path. This manuscript traces the evolution of the M&E landscape and discusses specific cases from the M&E industry where social media has been instrumental in hugely impacting the consumption of the M&E content.
... Behavioral segmentation using distance-based clustering and probability-based mixture models has been extensively studied in the fields including user targeting in marketing [12], online advertising [13] [14], personalizing content serving [15], identifying user behavior on social networks [16], and modeling shared interests of e-commerce users [17]. Applying segmentation in experimentation was also discussed in previous literature. ...
Preprint
Full-text available
Online controlled experimentation is widely adopted for evaluating new features in the rapid development cycle for web products and mobile applications. Measurement of the overall experiment sample is a common practice to quantify the overall treatment effect. In order to understand why the treatment effect occurs in a certain way, segmentation becomes a valuable approach to a finer analysis of experiment results. This paper introduces a framework for creating and utilizing user behavioral segments in online experimentation. By using the data of user engagement with individual product components as input, this method defines segments that are closely related to the features being evaluated in the product development cycle. With a real-world example, we demonstrate that the analysis with such behavioral segments offered deep, actionable insights that successfully informed product decision-making.
... Social networking websites, search engines and online ad networks make extensive use of personal data to target advertising to individual members of their audience [84,85,86]. For example, in "behavioral targeting", ads are served based on people's previous online activity and browsing behavior [87]. Empirical research has shown that ad targeting based on people's smallest expressions of preference, such as a single "like" on Facebook, can already result in an effect of mass psychological persuasion [88]. ...
Preprint
Full-text available
Even after decades of intensive research and public debates, the topic of data privacy remains surrounded by confusion and misinformation. Many people still struggle to grasp the importance of privacy, which has far-reaching consequences for social norms, jurisprudence, and legislation. Discussions on personal data misuse often revolve around a few popular talking points, such as targeted advertising or government surveillance, leading to an overly narrow view of the problem. Literature in the field tends to focus on specific aspects, such as the privacy threats posed by 'big data', while overlooking many other possible harms. To help broaden the perspective, this paper proposes a novel classification of the ways in which personal data can be used against people, richly illustrated with real world examples. Aside from offering a terminology to discuss the broad spectrum of personal data misuse in research and public discourse, our classification provides a foundation for consumer education and privacy impact assessments, helping to shed light on the risks involved with disclosing personal data.
... Historical data may include information about past preferences, topics of interest, or contents of online shopping carts. Current understanding of behavioral advertising is largely unchanged from its early definitions which characterized behavioral targeting as "the delivery of ads to targeted users based on information collected on each individual user's web search and browsing behaviors" (Yan et al. 2009). Such tracking is mentioned explicitly in contemporaneous definitions by industry participants such as Google (2021), Amazon (2021), and the wider Digital Advertising Alliance (DAA 2009) as well as by academic researchers, who emphasize the collection and tracking of previous online behavior (e.g., Boerman, Kruikemeier, and Borgesius 2017;Choi et al. 2020;Smit, Noort, and Voorveld 2014). ...
... Yan et al. [10] used user search queries and clicked pages respectively were used as the user behavior profiles to denote the users in vector space. The use of K-means algorithm was used and it was found that users who clicked the same advertisement will truly have similar behaviors on the Web. ...
Research
Full-text available
Online advertising markets are growing and continue to gain traction from websites, social networks, and mobile apps. But targeted advertising remains a topic of concern. How do we ensure targeted advertising? Online advertising uses different metrics to evaluate the success of a campaign, such as clicks on the advertisement, subscription to products, purchases of items, conversion, total site traffic, etc. Moreover, as the number of publishers and advertisers has risen to huge extent; the need for Real time bidding and prediction systems are critical for the success of the online advertising environment. In this present fourth industrial revolution, numerous techniques exist such as data, machine learning, artificial intelligence but in this paper we explore how data and machine learning has been used to improve online advertising.
... Therefore, it is necessary to ensure that the outputs from the systems are error-free to create the perception of usefulness. Since BTA usually results in more precise and effective advertisement delivery (Yan et al., 2009), it can be argued that consumer PU will be higher if the perception of OQ is high. As a result, we propose the following hypothesis: ...
Article
The study investigates the antecedents that affect consumers’ acceptance of behavioral targeting advertising (BTA) services by extending technology acceptance Model 2 (TAM2) with perceived risk. A two-stage PLS-SEM-artificial-neural-network (ANN) predictive analytic approach was adopted to analyze the collected data, of which PLS-SEM was first applied to test the hypotheses, followed by the ANN technique to detect the nonlinear effect on the model. A total of 475 usable self-administered questionnaires were collected, and the results showed that only the relationship between the image and perceived usefulness (PU) was not supported. As per Model B, the ranking of subjective norms (SN) and PU between the PLS-SEM and ANN model does not match each other, implying that hidden attributes may exist in affecting the role of SN and PU under the practical context of which the relationship between variables may not fully be explained by a linear perspective. The finding is beneficial for advertising practitioners and software developers who wish to optimize BTA results. Theoretically, the study extends TAM2 in the context of advertising, which is a neglected research area. Methodologically, the study is the first to apply TAM2 using the hybrid PLS-SEM-ANN in the context of advertising.
... For most personalization processes and experiences, big data such as geolocation, source, buyer persona, or buyer status are made use of to draw inferences on customers' needs and preferences [5]. Marketing messages are thereby adapted to customers using provided information by utilizing different media channels and tactics. ...
Article
Full-text available
With a mass of customer data at our fingertips and the ability to use it to individualize promotion strategies, marketing communications, and product offerings, marketing activities are becoming more and more tailored to the individual customer. However, as concerns about online privacy and the handling of personal data take on an ever-increasing significance, marketers must increasingly evaluate and adapt their personalization and data collection methods. As a result, there is an increasing demand to take a critical look at the collection of data for personalization processes from an ethical perspective and to consider implications for further initiatives to maintain consumers’ trust. This research study utilizes a systematic literature review approach to investigate the current state of knowledge on the tradeoff between personalization and customer privacy by synthesizing and integrating extant knowledge. From the results of the present study’s search process, 20 articles were selected and analyzed for this review. Findings emphasize the importance of strengthening consumer relationships by increasing consumer trust, loyalty, confidence, and emotional attachment through specific organizational activities. The adaptation of marketing-related activities can thereby create a competitive advantage for data-collecting companies, as consumer backlash and privacy concerns decrease, and the willingness to disclose data increases. The current research contributes to the field of marketing by reviewing the issue of increasing personalization at the cost of customer privacy and explores how the resulting ethical considerations may affect the future of marketing practices. It thereby serves to help marketeers to implement effective strategies to ensure customer relationships and the resulting willingness to disclose personal data for personalization processes.
Article
The European Union (EU)’s General Data Protection Regulation (GDPR), with its explicit consent requirement, may restrict the use of personal data and shake the foundations of online advertising. The ad industry predicted drastic loss of revenue from GDPR compliance and has been seeking alternative ways of targeting. Taking advantage of an event created by an ad publisher’s request for explicit consent from users with EU IP addresses, the authors find that for a publisher that uses a pay-per-click model, has the capacity to leverage both user behavior and webpage content information for advertising, and observes high consent rates, GDPR compliance leads to modest negative effects on ad performance, bid prices, and ad revenue. The changes in ad metrics can be explained by temporal variations in consent rates. The impact is most pronounced for travel and financial services advertisers and least pronounced for retail and consumer packaged goods advertisers. The authors further find that webpage context can compensate for the loss of access to users’ personal data, as the GDPR’s negative impact is less pronounced when ads are posted on webpages presenting relevant content. The results suggest that publishers and advertisers should leverage webpage-content-based targeting in a post-GDPR world.
Article
Targeted advertising can harm vulnerable groups when it targets individuals' personal and psychological vulnerabilities. We focus on how targeted weight-loss advertisements harm people with histories of disordered eating. We identify three features of targeted advertising that cause harm: the persistence of personal data that can expose vulnerabilities, over-simplifying algorithmic relevancy models, and design patterns encouraging engagement that can facilitate unhealthy behavior. Through a series of semi-structured interviews with individuals with histories of unhealthy body stigma, dieting, and disordered eating, we found that targeted weight-loss ads reinforced low self-esteem and deepened pre-existing anxieties around food and exercise. At the same time, we observed that targeted individuals demonstrated agency and resistance against distressing ads. Drawing on scholarship in postcolonial environmental studies, we use the concept of slow violence to articulate how online targeted advertising inflicts harms that may not be immediately identifiable. CAUTION: This paper includes media that could be triggering, particularly to people with an eating disorder. Please use caution when reading, printing, or disseminating this paper.
Article
Commerce has been moved to online platforms, and purchasing decisions have begun to be made online without face-to-face interaction with digitalization. The digitalization of interaction eliminates the product-specific information asymmetry between sellers and consumers in favor of consumers. On the other hand, the digital footprints left by individuals in digital environments and the internet's ability to offer personalized content independent of time and place have led to the emergence of a new type of information asymmetry. The processing of consumer data has made it possible to develop digital nudge messages and content containing cognitive biases and thus posing the risk of directing consumers to irrational consumption decisions. In this study, first of all, the product pages of e-commerce platforms were examined, and frequently used cognitive biases to direct consumers to irrational consumption were revealed. Subsequently, the frequency of consumers' decision-making under the influence of cognitive biases was determined by experiments with volunteer participants on an e-commerce page developed. In addition, the data of the participants and the rational behavior scores of the participants were obtained by questionnaire study, and the relationships between cognitive biases and personal data such as rationality score, age, gender, shopping frequency, and daily time spent on the internet were analyzed by binary logistic regression. As a result of the study, the factors affecting each cognitive bias were determined, and recommendations were made for further researches.
Article
Full-text available
Understanding the customer sentiment is very important when it comes to advertising. To appeal to their current and potential customers, a company must understand the market interests. Companies can segment their customers by using surveys and telemetry data to get to know the customer’s interests. One way of segmenting the customer is by grouping or clustering them according to their interests and behaviors. In this study, the k-prototypes clustering algorithm, which is an improved combination of k-means and k-modes algorithm, will be used to cluster a behavioral data that contains both numerical and categorical attribute, obtained from a survey conducted on teenagers into clusters of 4, 5, and 6. Each cluster will contain teenagers with certain behavior different from other clusters. And then by analyzing the results, advertisers will be able to define a profile that indicates their interests regarding the internet, social media and text messaging, effectively revealing the kind of ad that would be relatable for them.
Chapter
The concept of “personalized security nudges” promises to solve the contradictions between people’s heterogeneity and one-size-fits-all security nudges, whereas the psychological traits needed for personalization are not easy to obtain. To address the problem, we propose to leverage users’ behaviors logged by information systems, from which multiple behavioral features are extracted. A between-subjects lab experiment was conducted, during which participants’ behavioral features and responses to three famous security nudges (the so-called nudge effects) were logged. To test the feasibility of our proposal, we analyzed the relationships between the behavioral features with the nudge effects and discovered the significant moderation effects expected for all the three security nudges involved. The results indicate the feasibility of personalizing security nudges according to user behaviors, liberating the personalized security nudge schemes from the dependence on psychological scales.KeywordsNudgePersonalizationBehavioral features
Article
Within online advertising, there has been remarkable growth in display advertising. Our study focuses on the publisher’s allocation of its targeted advertising inventory – to this end, we employ the Hotelling model and use maximally representative allocation in our calculations. We then determine the supremum and infimum prices of a guaranteed contract allocated to advertisers. Rather significantly, our model provides a hybrid channel strategy for a publisher that allows the traditional method of targeted advertising and real-time bidding to coexist. We thus demonstrate the difference in each strategy’s outcome. The study reveals that the revenue of the publisher increases when it employs a hybrid channel strategy. Further, advertisers can bid for real-time bidding impressions in pursuit of high user conversion and simultaneously purchase a guaranteed contract in advance to reduce risk under conditions of sufficient funding.
Article
Under the pressure of surging advertising costs and fierce market competition, a growing number of online merchants are trying to implement new advertising strategies to improve advertising efficiency and increase sales. In the context of the widespread acceptance of online platform membership, many merchants who lack the ability of Big Data choose to cooperate with platforms to implement targeted advertising strategy (TA strategy). Under this strategy, merchants can separately push different ads to the members and nonmembers, which breaks the past practice of pushing unified ads to the whole market under mass advertising strategy (MA strategy). The platform can obtain the member-only pricing right from the cooperation to enhance the attractiveness of its membership system and charge a higher advertising fee. This article discusses the interaction between the merchant and the platform and obtains the optimal decisions under MA and TA, respectively. Then, we compare the profit performance, advertising efficiency, and product sales between the two strategies. The result shows that the merchant will adopt TA when the gap in charging standards of the two strategies is narrow. Furthermore, when the gap is moderate, the dominance of TA relies on the small proportion of the member group. In addition, we find that the merchant prefers to reduce (increase) the advertising level to members (nonmembers) under TA, in comparison with MA. Through demand analysis under two strategies, a counterintuitive result shows that, in the TA scenario, the demand for products generated by the member group will decrease with the rise of the matching degree.
Article
As the number and variety of online channels increase, businesses and advertisers must realize efficient marketing strategies and marketing mix. Customers' online channel usage patterns can be heterogeneous depending on various variables. If they understand this behavior correctly and use them for channel inflow or advertisement toward customers, they can establish an effective marketing strategy. In this paper, we utilize sequential pattern mining to find homogeneous patterns in customer's conversion paths by using conversion path data of customers who purchase diapers and find related insights. Furthermore, by analyzing the customer's channel movement pattern by dividing it into demographic information, we examine how demographic factors affect customer's online channel usage behavior.
Article
Full-text available
Günümüzde, teknolojinin hızla gelişimi ve internetin günlük yaşamın her alanına derinden nüfuz etmesi tüketici davranışında büyük değişimlere sebep olmaktadır. İşletmeler, bir bireyin durumuna ve ihtiyaçlarına göre kişiselleştirilmiş içerik sağlayarak genel müşteri deneyimini geliştirmeye odaklanmaktadır. Bağlamsal pazarlama, doğru bilgileri, doğru zamanda ve doğru kişilere ulaştırmaya dayanan etkili bir pazarlama aracıdır. Bağlamsal pazarlama işletmelere, uygun maliyetle daha başarılı hedefleme, potansiyel müşterileri aktif müşterilere dönüştürmek, satışları ve marka farkındalığını arttırmak, geleneksel reklamcılığın tüketiciyi kesintiye uğratan konseptini yok etmek gibi çok sayıda fayda sağlamaktadır. Bununla birlikte, literatürde bağlamsal pazarlama hakkında yapılan çalışmalar oldukça kısıtlıdır ve bu alanda çok daha fazla araştırmaya ihtiyaç bulunmaktadır. Bu çalışma, bağlamsal pazarlama ilgili geniş bir teorik çerçeve oluşturmaya odaklanmaktadır. Ayrıca çalışma, konuyla ilgili güncel istatistikler ve bağlamsal pazarlama stratejilerini de inceleyerek ilgili kararları alırken işletme yöneticilerine yardımcı olmayı amaçlamaktadır.
Conference Paper
Full-text available
Contextual advertising or Context Match (CM) refers to the placement of commercial textual advertisements within the content of a generic web page, while Sponsored Search (SS) advertising consists in placing ads on result pages from a web search engine, with ads driven by the originating query. In CM there is usually an intermediary commercial ad-network entity in charge of optimizing the ad selection with the twin goal of increasing revenue (shared between the publisher and the ad-network) and improving the user experience. With these goals in mind it is preferable to have ads relevant to the page content, rather than generic ads. The SS market developed quicker than the CM market, and most textual ads are still characterized by "bid phrases" representing those queries where the advertisers would like to have their ad displayed. Hence, the first technologies for CM have relied on previous solutions for SS, by simply extracting one or more phrases from the given page con- tent, and displaying ads corresponding to searches on these phrases, in a purely syntactic approach. However, due to the vagaries of phrase extraction, and the lack of context, this approach leads to many irrelevant ads. To overcome this problem, we propose a system for contextual ad matching based on a combination of semantic and syntactic features.
Article
In k\hbox{-}{\rm{means}} clustering, we are given a set of n data points in d\hbox{-}{\rm{dimensional}} space {\bf{R}}^d and an integer k and the problem is to determine a set of k points in {\bf{R}}^d, called centers, so as to minimize the mean squared distance from each data point to its nearest center. A popular heuristic for k\hbox{-}{\rm{means}} clustering is Lloyd's algorithm. In this paper, we present a simple and efficient implementation of Lloyd's k\hbox{-}{\rm{means}} clustering algorithm, which we call the filtering algorithm. This algorithm is easy to implement, requiring a kd-tree as the only major data structure. We establish the practical efficiency of the filtering algorithm in two ways. First, we present a data-sensitive analysis of the algorithm's running time, which shows that the algorithm runs faster as the separation between clusters increases. Second, we present a number of empirical studies both on synthetically generated data and on real data sets from applications in color quantization, data compression, and image segmentation.
Article
We present a very brief history of the origins of sponsored search, which is the presentation of text advertisements in response to a user's search query.
Article
The experimental evidence accumulated over the past 20 years indicates that text indexing systems based on the assignment of appropriately weighted single terms produce retrieval results that are superior to those obtainable with other more elaborate text representations. These results depend crucially on the choice of effective termweighting systems. This article summarizes the insights gained in automatic term weighting, and provides baseline single-term-indexing models with which other more elaborate content analysis procedures can be compared.
Conference Paper
Several approaches to collaborative filtering have been stud- ied but seldom have studies been reported for large (several million users and items) and dynamic (the underlying item set is continually changing) settings. In this paper we de- scribe our approach to collaborative filtering for generating personalized recommendations for users of Google News. We generate recommendations using three approaches: collabo- rative filtering using MinHash clustering, Probabilistic La- tent Semantic Indexing (PLSI), and covisitation counts. We combine recommendations from different algorithms using a linear model. Our approach is content agnostic and con- sequently domain independent, making it easily adaptable for other applications and languages with minimal effort. This paper will describe our algorithms and system setup in detail, and report results of running the recommendations engine on Google News.
Article
Information retrieval studies that involve searching the Internet or marking phrases usually lack a well-defined number of negative cases. This prevents the use of traditional interrater reliability metrics like the kappa statistic to assess the quality of expert-generated gold standards. Such studies often quantify system performance as precision, recall, and F-measure, or as agreement. It can be shown that the average F-measure among pairs of experts is numerically identical to the average positive specific agreement among experts and that kappa approaches these measures as the number of negative cases grows large. Positive specific agreement-or the equivalent F-measure-may be an appropriate way to quantify interrater reliability and therefore to assess the reliability of a gold standard in these studies.
Article
The nearest neighbor problem is the following: Given a set of n points P = fp 1 ; : : : ; png in some metric space X, preprocess P so as to efficiently answer queries which require finding the point in P closest to a query point q 2 X. We focus on the particularly interesting case of the d-dimensional Euclidean space where X = ! d under some l p norm. Despite decades of effort, the current solutions are far from satisfactory; in fact, for large d, in theory or in practice, they provide little improvement over the brute-force algorithm which compares the query point to each data point. Of late, there has been some interest in the approximate nearest neighbors problem, which is: Find a point p 2 P that is an ffl-approximate nearest neighbor of the query q in that for all p 0 2 P , d(p; q) (1 + ffl)d(p 0 ; q). We present two algorithmic results for the approximate version that significantly improve the known bounds: (a) preprocessing cost polynomial in n and d, and a trul...