Recommender Systems: Sources of Knowledge and Evaluation Metrics

Parra, Denis; Sahebi, Shaghayegh

doi:10.1007/978-3-642-33326-2_7

Cited by 40 publications

(31 citation statements)

References 110 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…One such challenge is incorporating Human Factors in order to increase user acceptance of the systems and the items recommended. Historically, the focus on recommender systems' research has been on improving the algorithms' predictive accuracy (Parra & Sahebi, 2013), but as McNee et al (McNee, Riedl, & Konstan, 2006b) highlighted in the paper "Being accurate is not enough: how accuracy metrics have hurt recommender systems," accuracy does not always correlate with a good user experience, making the study of recommender interfaces one of the areas in need for improvement.…”

Section: Introductionmentioning

confidence: 99%

User-controllable personalization: A case study with SetFusion

Parra

Brusilovsky

2015

International Journal of Human-Computer Studies

Self Cite

View full text Add to dashboard Cite

Abstract.In this research we investigated the role of user controllability on personalized systems by implementing and studying a novel interactive recommender interface, SetFusion. We examined whether allowing the user to control the process of fusing or integrating different algorithms (i.e., different sources of relevance) resulted in increased engagement and a better user experience. The essential contribution of this research stems from the results of a user study (N=40) of controllability in a scenario where users could fuse different recommendation approaches, with the possibility of inspecting and filtering the items recommended. First, we introduce an interactive Venn diagram visualization, which combined with sliders, can provide an efficient visual paradigm for information filtering. Second, we provide a three-fold evaluation of the user experience: objective metrics, subjective user perception, and behavioral measures. Through the analysis of these metrics, we confirmed results from recent studies, such as the effect of trusting propensity on accepting the recommendations and also unveiled the importance of features such as being a native speaker. Our results present several implications for the design and implementation of user-controllable personalized systems.

show abstract

Section: Introductionmentioning

confidence: 99%

User-controllable personalization: A case study with SetFusion

Parra

Brusilovsky

2015

International Journal of Human-Computer Studies

Self Cite

View full text Add to dashboard Cite

show abstract

“…Normalized Discounted Cumulative Gain (nDCG@k) is a rankingdependent metric that not only measures how many products can be correctly predicted but also takes the position of the products in the recommended list with length k into account. The nDCG metric is based on the Discounted Cummulative Gain (DCG@k) which is given by [19]:…”

Section: Evaluation Methods and Metricsmentioning

confidence: 99%

Utilizing Online Social Network and Location-Based Data to Recommend Products and Categories in Online Marketplaces

Lacic

Kowald

Eberhard

et al. 2014

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. Recent research has unveiled the importance of online social networks for improving the quality of recommender systems and encouraged the research community to investigate better ways of exploiting the social information for recommendations. To contribute to this sparse field of research, in this paper we exploit users' interactions along three data sources (marketplace, social network and location-based) to assess their performance in a barely studied domain: recommending products and domains of interests (i.e., product categories) to people in an online marketplace environment. To that end we defined sets of content-and network-based user similarity features for each data source and studied them isolated using an user-based Collaborative Filtering (CF) approach and in combination via a hybrid recommender algorithm, to assess which one provides the best recommendation performance. Interestingly, in our experiments conducted on a rich dataset collected from SecondLife, a popular online virtual world, we found that recommenders relying on user similarity features obtained from the social network data clearly yielded the best results in terms of accuracy in case of predicting products, whereas the features obtained from the marketplace and locationbased data sources also obtained very good results in case of predicting categories. This finding indicates that all three types of data sources are important and should be taken into account depending on the level of specialization of the recommendation task.

show abstract

“…They said that the ideal scenario for them would be if the system told them, in advance, what they should be reading. In this regard, content-based recommender systems provide recommendations to users based on comparing items or products to the items in which the user has shown an interest [83], and collaborative filtering technology, as well as other forms of recommendation, automates the process of sharing opinions on the relevance and quality of information [84]. The advanced capabilities of MCC make it possible to apply machine learning algorithms to the information being generated by the feedback and evaluations, allowing the generation of suggestions and recommendations.…”

Section: (D) Lesson Evaluation (R6) and Future Recommendations (R8)mentioning

confidence: 99%

A User-Centered Mobile Cloud Computing Platform for Improving Knowledge Management in Small-to-Medium Enterprises in the Chilean Construction Industry

et al. 2018

View full text Add to dashboard Cite

Knowledge management (KM) is a key element for the development of small-to-medium enterprises (SMEs) in the construction industry. This is particularly relevant in Chile, where this industry is composed almost entirely of SMEs. Although various KM system proposals can be found in the literature, they are not suitable for SMEs, due to usability problems, budget constraints, and time and connectivity issues. Mobile Cloud Computing (MCC) systems offer several advantages to construction SMEs, but they have not yet been exploited to address KM needs. Therefore, this research is aimed at the development of a MCC-based KM platform to manage lessons learned in different construction projects of SMEs, through an iterative and user-centered methodology. Usability and quality evaluations of the proposed platform show that MCC is a feasible and attractive option to address the KM issues in SMEs of the Chilean construction industry, since it is possible to consider both technical and usability requirements.

show abstract

Recommender Systems: Sources of Knowledge and Evaluation Metrics

Cited by 40 publications

References 110 publications

User-controllable personalization: A case study with SetFusion

User-controllable personalization: A case study with SetFusion

Utilizing Online Social Network and Location-Based Data to Recommend Products and Categories in Online Marketplaces

A User-Centered Mobile Cloud Computing Platform for Improving Knowledge Management in Small-to-Medium Enterprises in the Chilean Construction Industry

Contact Info

Product

Resources

About