Chunyan Hou's research while affiliated with Tianjin University of Technology and other places

Publications (13)

Article
The explosive growth of fake news on social media has drawn great concern both from industrial and academic communities. There has been an increasing demand for fake news detection due to its detrimental effects. Generally, news content is condensed and full of knowledge entities. However, existing methods usually focus on the textual contents and...
Article
Rumors often yield adverse societal and economic impacts. Therefore, rumor detection has attracted a surge of research interests. Existing methods mainly focus on finding clues from textual contents, which is not quite effective as rumors can be intentionally manipulated. Recent studies have demonstrated that the propagation structure of rumors can...
Article
In the era of e-commerce, purchase behavior prediction is one of the most important issues to promote both online companies' sales and the consumers' experience. The previous researches usually use the feature engineering and ensemble machine learning algorithms for the prediction. The performance really depends on designed features and the scalabi...
Article
System scenarios that derived from system design specification play an important role in the reliability engineering of componentbased software systems. Several scenario-based approaches have been proposed to predict the reliability of a system at the design time, most of them adopt flat construction of scenarios, which doesn't conform to software...
Article
In the era of e-commerce, purchase behavior prediction is one of the most important issues to promote both online companies' sales and the consumers' experience. The previous researches usually use traditional features based on the statistics and temporal dynamics of items. Those features lead to the loss of detailed items' information. In this stu...
Article
Purchase behavior prediction is one of the most important issues to promote both e-commerce companies' sales and the consumers' satisfaction. The prediction usually uses features based on the statistics of items. This kind of features can lead to the loss of detailed information of items. While all items are included, a large number of features has...
Article
Purchase behavior prediction is one of the most important issues for the precision marketing of e-commerce companies. This Letter presents our solution to the purchase behavior prediction problem in E-commerce, specifically the task of Big Data Contest of China Computer Federation in 2014. The goal of this task is to predict which users will have t...
Article
Dynamic Spectrum Access (DSA) network grows rapidly in recent years. It is proposed to solve the problem of reasonable utilization of wireless spectrum resources. DSA is a new spectrum sharing paradigm which takes advantage of spectrum holes to ease the spectrum shortage problem and improve the spectrum utilization. However, due to the frequent spe...
Article
Recommendation systems have been widely used in E-commerce sites, social media and etc. An important recommendation task is to predict items that a user will perform actions on with users' historical data, which is called top-K recommendation. Recently, there is huge amount of emerging items which are divided into a variety of categories and r...
Article
With the rise of component-based software development, its reliability has attracted much attention from both academic and industry communities. Component-based software development focuses on architecture design, and thus it is important for reliability analysis to emphasize software architecture. Existing approaches to architecture-based software...

Citations

... Second, a brief textual material frequently carries a rich background knowledge. Existing natural language comprehension models frequently fail to cover such a vast scope of knowledge (Dun et al., 2021), making it challenging to comprehend the exact content of news articles and resulting in a decline in performance. Previous studies (Liu and Wu, 2018) have found that the rumor propagates differently than real news, which suggests that the propagation network of news on social networks can be used to detect rumors. ...
... However, their feature recognition work is driven by data rather than theories [25], which causes the existing approaches to only perform well on specific datasets. Furthermore, most current works focus on the linguistic features [26][27][28], which may be deliberately intended for specific purposes, e.g., deceptive language features can be injected to bypass the text-based rumor detection model [29]. Thus, the fragility of linguistic features prompts researchers to start looking for more reliable features, e.g., propagation features (such as propagation range, i.e., the size of the retweet network). ...
... Nevertheless, several studies reveal how consumer behavior prediction using data mining techniques was successful in developing association rule mining models to predict customer behavior managing to export new sets of rules value (Orogun and Bukola,2019). Compared to similar consumer behavior prediction attempts there is one which demonstrates an alternative use tree-based feature transformation and machine learning algorithms, instead of using ensemble algorithms, a simple algorithm is used to predict purchase behavior based on transformed features (Hou et al., 2018). Another approach in predicting consumers' behavior this time using random forest data mining technique proved to be effective since it has managed to succeed high accurate results (Valecha et al., 2018). ...
... Many studies have been carried out based on these behavior models to predict the reliability at early design stages [7,[13][14][15][16][17][18]. However, the current models need to pay more attention to scenario combination mechanism and model scalability for larger systems [8][9][10]19]. The scalability problem of the model is related to system behavior modeling. ...
... e proposed algorithm can be integrated into management practice as well: the task of predicting future product demand can be completed by connecting to the enterprise's database, then reading a large amount of data before conducting a series of data analyses and running the algorithm's internal processing mechanism [70,71]. When changes in product demand are updated and synchronized in real time, the effect of real-time prediction and allocation of goods can be achieved as well. is study was the first to combine the problems of order demand forecasting and multiobjective space allocation, giving reference to both order management and space allocation for warehouses. ...
... Table 1 brings single task proposals (prediction of one outcome), while Table 2 provides multi-task proposals (prediction of multiple outcomes). [15] x Feature engineering for clickstream [16] x x Association rules for fast predictions [17] x x Feature engineering for popular products [18] x x x Feature engineering from customer search [19] x Benchmark over multiple online shops [20] x x Feature engineering for multiple products [21] x x Feature engineering with graph metrics [22] x x K-Nearest Neighbor for fast predictions [23] x x Feature engineering with motifs in single sessions [24] x x PC Prediction over multiple online visits [25] x x DLC Feature learning for automatic feature construction [26] x x x Real-time predictions in single visits [27] Product ...
... Only a small number of active customers become repeat customers and purchase again. ese are called stable customers [3]. Only a few of the stable customers increase their purchases and deepen their relationship with the site, and these people become the "best customers." ...
... In order to solve this problem, Hou et al. [18] proposed extension to the work in [11] by separately modeling and evaluating each scenario to avoid state space explosion and reduce computational complexity. However, the approach analyzed case scenarios directly, and didn't consider the dependencies between different kinds of scenarios. ...