Conference Paper

Demographic recommendations for WEITBLICK, an assistance system for elderly

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

Abstract

This paper evaluates the possible usage of demographic recommender systems for an assistance system called WEITBLICK. The aim of WEITBLICK is to provide elderly with information about services from the areas care, health, recreation, household, etc. Three types of demographic recommender systems are studied. All of them use linear predictors to make assumptions about unknown ratings of items by the users. The predictors are learned by gradient descent (GD), exponentiated gradient descent (EG), and exponentiated gradient descent with positive and negative weights (EG<sup>±</sup>). Using a data set from a user survey, it is shown that EG and EG<sup>±</sup> perform best. Furthermore, a way to reduce computing time while only trading in a reasonable amount of accuracy is explained. A discussion about the usage of the results for further research is provided.

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.

... Another proposed intervention recommendation approach was focused on a demographic recommender system for the elderly [15]. This recommender system focused on the demographic aspect to provide elderly with information about services of health, recreation, household, etc. ...
... The weightage was collected from the experts of each aspect in order to calculate the overall condition of the elderly. The weightages for each aspect are as follows: socialization (10), health (30), cognitive (15), physical (15), nutrition (10), spiritual (10) and environment (10). Below is the formula to calculate the weightage for each aspect based on the assessment data collected from the elderly. ...
... The weightage was collected from the experts of each aspect in order to calculate the overall condition of the elderly. The weightages for each aspect are as follows: socialization (10), health (30), cognitive (15), physical (15), nutrition (10), spiritual (10) and environment (10). Below is the formula to calculate the weightage for each aspect based on the assessment data collected from the elderly. ...
... Most of recommended systems are based on content-based (CB), or collaborative filtering (CF) algorithm. The latter is more commonly used, which creates recommended results based on the preference similarity between the users [6,10,12,14]. ...
... Generally, the recommendation method [5][6][7][8] of the data mining [4] consists of the followings: content-based method, demographic method, and collaborative filtering method. Content-based method is word frequency method. ...
... Demographic method is what predicts object preference of a specific user by calculating object preferences of other users using profiles of people such as job, gender, and age, etc., [6,9,10,11]. Collaborative filtering (CF) method extracts objects by discerning other users having similar patterns to the client recommendation will be given to and using information on object evaluation of them based on their preference histories [12][13][14][15][16]. ...
Article
Popularization of smartphone allows many people to be able to share information, and easily get information they want. And then, app store of smartphone plays an important role to make users get information and make transactions they want easily. With the increase of the number of apps traded at the app store, it is difficult for users to find appropriate apps they want. Commonly, app store recommends apps users want through key words posted by users. But, this method has limits, and may not be satisfactory to users. This paper proposes a method where app store recommends apps to users by classifying apps used by users’ friends by category and degree of satisfaction and recommending apps which are likely to be used. The experiment to test satisfaction of users showed that the method proposed in this paper increased the degree of satisfaction of users by over 21%.
... For example, [30] proposed a contextaware-based location recommender system that can monitor the location of the elderly and deliver appropriate location recommendations by considering context. On the other hand, Stiller et al. [31] used demographic attributes in order to provide personalised information about available services in the surroundings of elderly users, while Kitamura et al. [32] proposed a system for recommending social services to elderly by classifying social services and subjective experiences among the elderly using the World Health Organization's (WHO) International Classification of Functioning, Disability and Health (ICF) [33]. ...
Full-text available
Article
During the recent years, several new Information and Communication Technology solutions have been developed in order to meet the increasing needs of elderly with cognitive impairments and support their autonomous living. Most of these solutions follow a human-centred paradigm that aims to provide users with personalised services according to their needs by also ensuring their safety with mechanisms that can automatically trigger appropriate actions in situations where there may be a risk for an elderly. The present paper presents a hybrid matchmaking approach that uses efficiently both a rule-based and a statistical matchmaker in order to (a) propose ambient assisted living services to the end-users, based on their role, status and context of use and (b) identify and resolve problematic cases by automatically selecting the most proper set of services to be called in a single or combined manner.
... 객체 추출에 이러한 인구통계학적 정보를 활용할 경우에는 근접 사용자의 객체 선호도를 계산하여, 특정 사용자의 객체 선호도를 예측하 는 방식을 따른다. 이러한 인구통계학적 기법은 객체 집단을 쉽게 구별할 수 있기 때문에 대량 의 객체를 포함한 영역에서 효과적인 기법이다[7][8]. 인구통계학적 기법을 적용하기 위해서는 사용자의 신상정보가 필요하다. ...
Article
The remarkable development of IT is contributed to popularization of smart phones, which in turn creates a new domain called app store. Smartphone apps have grown fast because they can be easily purchased through an app store. As the volume of apps traded in app stores is so huge that it is extremely hard for users to find the exact app they want. In general, an app store recommends an app to users based on the search words they entered. In terms of recommendation of app, this kind of content-based method is not effective. To increase accuracy in recommending app, this paper proposes a characteristic similarity-based app recommendation method. This method creates attributes on the app based on the related information such as genre, functionality and number of downloads and then compares them with the propensity to use the app. According to diverse simulations, the method proposed in this paper improved the performance of app recommendation by 33% in average, compared to the conventional method.
... Danach kommen in der zweiten Stufe mehrere Unterempfehlungssysteme parallel zum Einsatz, die jeweils verschiedene Kriterien als Bewertungsbasis nutzen. Sie unterscheiden sich damit je nach Nutzung des eigentlichen Empfehlungsansatzes nach kollaborativen, demografischen oder inhaltsbasierten Empfehlungssystemen (Stiller et al. 2010a, Stiller et al. 2010b, Stiller et al. 2009). Zudem können auch durch die Nutzung einer OSGi-Umgebung sehr einfach weitere Teilempfehlungssysteme in die modulare Architektur eingefügt werden. ...
Chapter
Technologiegetriebene, innovative Leistungsangebote können fundamentale Veränderungsprozesse in traditionellen Branchenstrukturen auslösen (vgl. Christensen et al. 1998; Gambardella/McGahan 2010; Gersch/Goeke 2007, Jacobides et al. 2006). Der Einsatz von Informations- und Kommunikationstechnik kann sich direkt auf die Art und Weise der Leistungserstellung und Koordination bestehender beziehungsweise neuer Wertschöpfungsketten und damit auf die Struktur arbeitsteiliger Leistungserstellungsprozesse sowie jeweils realisierter Leistungsangebote auswirken. Im Rahmen der Digitalisierung und Vernetzungen von Leistungsangeboten können bisher vorliegende Restriktionen und Begrenzungen bereits etablierter Angebote überwunden werden und so zu einer neuen Branchenarchitektur in bisherigen und/oder neuen Absatzmärkten führen (vgl. Schmid 2001; Shapiro/Varian 1998). Eine solche Entwicklung ist derzeit im Bereich des Gesundheitswesens erkennbar, insbesondere am Gesundheitsstandort „zu Hause“ (Fachinger/Henke 2010; Hilbert 2010). Am Beispiel Ambient Assisted Living (AAL) veranschaulicht dieser Beitrag, wie die Ausgestaltung bisheriger Angebote durch Anwendung neuer Technologien grundlegend verändert wird.
... For the adopted approach, recommender systems can be classified in the following categories (Burke 2007): (i) Content-based (CB) that considers past users' interests (Lops et al. 2011); (ii) Collaborative Filtering (CF), which realizes a knowledge-sharing environment to find people having similar interests for generating suggestions about items resulting unknown to the users (Breese et al. 1998;Su and Khoshgoftaar 2009); (iii) Demographic, to identify those recommender users belonging to the same demographic niche (Stiller et al. 2010); (iv) Knowledge-based, for inferring user's needs and references (Trewin 2000). A further category, commonly recognized as the most performing, is that of the hybrid systems (Burke 2002) which combine more approaches to promote mutual synergies and improve effectiveness and efficiency of the recommendation process. ...
Article
Recommender systems support users by generating potentially interesting suggestions about relevant products and information. The increasing attention towards such tools is witnessed by both the great number of powerful and sophisticated recommender algorithms developed in recent years and their adoption in many popular Web platforms. However, performances of recommender systems can be affected by many critical issues as for instance, over-specialization, attribute selection and scalability. To mitigate some of such negative effects, a hybrid recommender system, called Relevance Based Recommender, is proposed in this paper. It exploits individual measures of perceived relevance computed by each user for each instance of interest and, to obtain a better precision, also by considering the analogous measures computed by the other users for the same instances. Some experiments show the advantages introduced by this recommender when generating potentially attractive suggestions.
Article
In the last decades, individual urban traffic flows have increased all over the world with a consequent growth of road congestion and environmental pollution. In this context, car-pooling is an interesting car-based alternative to satisfy the individual mobility demand by optimizing the car loading factor with respect to the number of passengers, provided that all the participants share trip origin and destination in the same time slot. To make the system more appealing, this paper proposes an on-demand car-pooling service adopting variable fares, on the basis of trip length and number of participants. Multi-agent, reputation and recommender system technologies in synergy with a routing algorithm have been used to this aim. Experiments on simulated data proved the potentiality of the proposed approach.
Chapter
Ausgehend von aktuellen demografischen Daten und den daraus abgeleiteten Schätzungen wird deutlich, dass die ältere Bevölkerung relativ zu den jüngeren Altersgruppen wächst, wodurch auch mit einem steigenden Bedarf an Dienstleistungen für ältere Menschen zu rechnen ist. Zudem steigen durch verschiedene Faktoren begünstigt auch die Lebenserwartung und die längere Aufrechterhaltung der Gesundheit und Mobilität. Damit differenzieren sich Lebenslagen und -stile älterer Menschen stärker voneinander, wodurch diese keine homogene Gruppe mehr darstellen, denen gleiche Dienstleistungen sowohl im Pflege- als auch im Freizeitbereich angeboten werden können (Presse- und Informationsamt der Bundesregierung 2009).
Full-text available
Conference Paper
Um Senioren einen längeren Verbleib in der eigenen Häuslichkeit zu ermöglichen fehlt es mitunter nur an prinzipiell verfügbaren Informationen, die für den jeweiligen Senior jedoch nicht ermittelbar sind. Darüber hinaus können Angebote und Dienstleistungen häufig aufgrund fehlender Mobilität oder aus mangelndem Sicherheitsgefühl heraus nicht wahrgenommen werden. Diese behebbaren Defizite überbrückt die Informationsplattform Weitblick, indem sie die notwendigen Informationen bereitstellt, individuell aufbereitet und an die jeweiligen Wünsche und Befähigungen angepasst darstellt. Durch zusätzliche Monitoringfunktionen und das Angebot durchgängiger Planung von der Information bis zur Realisierung der Wahrnehmung von Angeboten wird sowohl für den Senior als auch den Betreiber ein Mehrwert geschaffen.
Full-text available
Article
Recommender systems represent user preferences for the purpose of suggesting items to purchase or examine. They have become fundamental applications in electronic commerce and information access, providing suggestions that effectively prune large information spaces so that users are directed toward those items that best meet their needs and preferences. A variety of techniques have been proposed for performing recommendation, including content-based, collaborative, knowledge-based and other techniques. To improve performance, these methods have sometimes been combined in hybrid recommenders. This paper surveys the landscape of actual and possible hybrid recommenders, and introduces a novel hybrid, EntreeC, a system that combines knowledge-based recommendation and collaborative filtering to recommend restaurants. Further, we show that semantic ratings obtained from the knowledge-based part of the system enhance the effectiveness of collaborative filtering.
Full-text available
Conference Paper
The changing demography requires new kinds of support for elderly people. The project WEITBLICK aims to give seniors assistance to gather information about several services and their providers, relaying the access to such services and offer them in an individualized manner. To determine the requirements of elderly users a broad analysis will be performed in four stages. To fulfill the aims of the project the system has two principles incorporated: the service relay can be triggered by the users’ former activities or by the users actively themselves. The base for both is a database with user and service profiles.
Full-text available
Article
Recently, Artificial Intelligence techniques have proved useful in helping users to handle the large amount of information on the Internet. The idea of personalized search engines, intelligent software agents, and recommender systems has been widely accepted among users who require assistance in searching, sorting, classifying, filtering and sharing this vast quantity of information. In this paper, we present a state-of-the-art taxonomy of intelligent recommender agents on the Internet. We have analyzed 37 different systems and their references and have sorted them into a list of 8 basic dimensions. These dimensions are then used to establish a taxonomy under which the systems analyzed are classified. Finally, we conclude this paper with a cross-dimensional analysis with the aim of providing a starting point for researchers to construct their own recommender system.
Full-text available
Article
Recommender systems have been evaluated in many, often incomparable, ways. In this article, we review the key decisions in evaluating collaborative filtering recommender systems: the user tasks being evaluated, the types of analysis and datasets being used, the ways in which prediction quality is measured, the evaluation of prediction attributes other than quality, and the user-based evaluation of the system as a whole. In addition to reviewing the evaluation strategies used by prior researchers, we present empirical results from the analysis of various accuracy metrics on one content domain where all the tested metrics collapsed roughly into three equivalence classes. Metrics within each equivalency class were strongly correlated, while metrics from different equivalency classes were uncorrelated.
Full-text available
Article
In this paper we explore how two existing collaborative flltering algo- rithms can be enhanced by the calculation of demographic correlations among the members of user or item neighborhoods. Experiments are executed to evaluate the performance of the proposed approach. Their results show that demographic data can, in some cases, lead to the generation of more accurate predictions.
Article
This paper presents an overview of the field of recommender systems and describes the current generation of recommendation methods that are usually classified into the following three main categories: content-based, collaborative, and hybrid recommendation approaches. This paper also describes various limitations of current recommendation methods and discusses possible extensions that can improve recommendation capabilities and make recommender systems applicable to an even broader range of applications. These extensions include, among others, an improvement of understanding of users and items, incorporation of the contextual information into the recommendation process, support for multicriteria ratings, and a provision of more flexible and less intrusive types of recommendations.
Conference Paper
This paper studies the ramp up behaviour of recommender systems that incorporate spatial relationships. Two previously proposed algorithms are compared regarding their recommendation quality in situations with only little data available. Additionally a third algorithm is proposed which can be used to improve recommendation quality during the ramp-upphase. It uses approaches from item-based recommender systems to gather knowledge from an already established recommender system and transports it to a new recommender system.
Conference Paper
Recommender systems aggregate information about users and items to be recommended to generate adequate recommendations. This paper proposes two approaches to include information about spatial releationships of users and items in order to improve the quality of recommendations. The two approaches are compared with non-spatial recommendation using a set of evaluation metrics. Special ramp-up problems, occurring when including spatial information into recommender systems and different application areas for spatial recommender systems are discussed.
Conference Paper
In this paper, we apply stacking, an ensemble learning method, to the problem of building hybrid recommendation systems. We also introduce the novel idea of using runtime metrics which represent properties of the input users/items as additional meta-features, allowing us to combine component recommendation engines at runtime based on user/item characteristics. In our system, component engines are level-1 predictors, and a level-2 predictor is learned to generate the final prediction of the hybrid system. The input features of the level-2 predictor are predictions from component engines and the runtime metrics. Experimental results show that our system outperforms each single component engine as well as a static hybrid system. Our method has the additional advantage of removing restrictions on component engines that can be employed; any engine applicable to the target recommendation task can be easily plugged into the system.
Article
We study the construction of prediction algorithms in a situation in which a learner faces a sequence of trials, with a prediction to be made in each, and the goal of the learner is to make few mistakes. We are interested in the case that the learner has reason to believe that one of some pool of known algorithms will perform well, but the learner does not know which one. A simple and effective method, based on weighted voting, is introduced for constructing a compound algorithm in such a circumstance. We call this method the Weighted Majority Algorithm. We show that this algorithm is robust in the presence of errors in the data. We discuss various versions of the Weighted Majority Algorithm and prove mistake bounds for them that are closely related to the mistake bounds of the best algorithms of the pool. For example, given a sequence of trials, if there is an algorithm in the pool A that makes at most m mistakes then the Weighted Majority Algorithm will make at most c(log jAj + m) mi...
Article
We discuss learning a profile of user interests for recommending information sources such as Web pages or news articles. We describe the types of information available to determine whether to recommend a particular page to a particular user. This information includes the content of the page, the ratings of the user on other pages and the contents of these pages, the ratings given to that page by other users and the ratings of these other users on other pages and demographic information about users. We describe how each type of information may be used individually and then discuss an approach to combining recommendations from multiple sources. We illustrate each approach and the combined approach in the context of recommending restaurants. 2 1 Introduction Users are constantly confronted with situations in which they have many options to choose from and need assistance exploring or winnowing down the possibilities. Internet Search Engines commonly find thousands of potentially releva...
Article
We consider two algorithms for on-line prediction based on a linear model. The algorithms are the well-known gradient descent (GD) algorithm and a new algorithm, which we call EG(+/-). They both maintain a weight vector using simple updates. For the GD algorithm, the update is based on subtracting the gradient of the squared error made on a prediction. The EG(+/-) algorithm uses the components of the gradient in the exponents of factors that are used in updating the weight vector multiplicatively. We present worst-case loss bounds for EG(+/-) and compare them to previously known bounds for the GD algorithm. The bounds suggest that the losses of the algorithms are in general incomparable, but EG(+/-) has a much smaller loss if only few components of the input are relevant for the predictions. We have performed experiments which show that our worst-case upper bounds are quite tight already on simple artificial data. (C) 1997 Academic Press.
AAL in der alternden Gesellschaft – Anforderungen, Akzeptanz und Perspektiven – Analyse und Planungs-hilfen
  • S Meyer
  • H Mollenkopf
  • Eds
S. Meyer and H. Mollenkopf, Eds., AAL in der alternden Gesellschaft – Anforderungen, Akzeptanz und Perspektiven – Analyse und Planungs-hilfen. VDE Verlag GmbH, Berlin, Offenbach, 2010.
Concept of a Communication Middleware for Ambient Assisted Living Environments
  • K Renhak
  • M Federspiel
  • J Seitz
  • H.-P Schade
Personalisation of Location-Based Services using Content-Based Recommender Systems
  • C Stiller
  • F Roβ