Cold-start news recommendation with domain-dependent browse graph

Trevisiol, Michele; Aiello, Luca Maria; Schifanella, Rossano; Jaimes, Alejandro

doi:10.1145/2645710.2645726

Cited by 38 publications

(35 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Popular news portals, such as Google News [15], Yahoo! News [62], The New York Times [61], Washington Post [54] [10], among others have gained increasing attention from a massive amount of online news readers.…”

Section: News Recommender Systemsmentioning

confidence: 99%

News Session-Based Recommendations using Deep Neural Networks

Moreira

Ferreira²,

Cunha

2018

Proceedings of the 3rd Workshop on Deep Learning for Recommender Systems

View full text Add to dashboard Cite

News recommender systems are aimed to personalize users experiences and help them to discover relevant articles from a large and dynamic search space. Therefore, news domain is a challenging scenario for recommendations, due to its sparse user profiling, fast growing number of items, accelerated item's value decay, and users preferences dynamic shift.Some promising results have been recently achieved by the usage of Deep Learning techniques on Recommender Systems, specially for item's feature extraction and for session-based recommendations with Recurrent Neural Networks.In this paper, it is proposed an instantiation of the CHAMELEON -a Deep Learning Meta-Architecture for News Recommender Systems. This architecture is composed of two modules, the first responsible to learn news articles representations, based on their text and metadata, and the second module aimed to provide session-based recommendations using Recurrent Neural Networks.The recommendation task addressed in this work is next-item prediction for users sessions: "what is the next most likely article a user might read in a session?"Users sessions context is leveraged by the architecture to provide additional information in such extreme cold-start scenario of news recommendation. Users' behavior and item features are both merged in an hybrid recommendation approach.A temporal offline evaluation method is also proposed as a complementary contribution, for a more realistic evaluation of such task, considering dynamic factors that affect global readership interests like popularity, recency, and seasonality.Experiments with an extensive number of session-based recommendation methods were performed and the proposed instantiation of CHAMELEON meta-architecture obtained a significant relative improvement in top-n accuracy and ranking metrics (10% on Hit Rate and 13% on MRR) over the best benchmark methods. * Also with Brazilian Aeronautics Institute of Technology -ITA.

show abstract

Section: News Recommender Systemsmentioning

confidence: 99%

News Session-Based Recommendations using Deep Neural Networks

Moreira

Ferreira²,

Cunha

2018

Proceedings of the 3rd Workshop on Deep Learning for Recommender Systems

View full text Add to dashboard Cite

show abstract

“…Indeed, studying the users browsing patterns, Liu et al [20] showed that more recent clicks have a considerably higher value to predict future actions than the historical browsing record. Finally, Trevisiol et al [27] exploited the BrowseGraph in order to build some user models in the news domain, and recommend the next article the user is going to visit. They introduced the concept of ReferrerGraph, which is a BrowseGraph built with sessions that are generated by the same referrer domain.…”

Section: Browsegraphmentioning

confidence: 99%

“…For the purpose of this study, we took a sample of Yahoo News network's 1 user-anonymized log data collected in 2013. The dataset used in this work has been extracted from the data built in [27], that was used with the purpose to study the news consumption with respect to the Referrer URL. In this section we summarize how we built the dataset and the graphs, but the reader may refer to the aforementioned paper for further details.…”

Section: Datasetmentioning

confidence: 99%

“…Second, we learn that it is possible to infer the user domain of origin from the very first browsing steps. This capability enables several types of services, including user profiling [12], web site optimization [31], user engagement estimation [18], and cold-start recommendation [27], even when the referrer URL is not available (e.g.when the user comes from mobile social media applications or URL shortening services). Once we show that the subgraphs are different enough, we proceed to perform more involved experiments that we call "Growing Rings".…”

Section: Introductionmentioning

confidence: 99%

“…• We study the LRP on a large-scale BrowseGraph built from a very popular news website. To the best of our knowledge we are the first to tackle this problem on the increasingly popular BrowseGraph [27,28,12,22]. We present an analysis of the convergence of the PageRank on the local graph to the global one, by incrementally expanding the local graph in a snowingreen fashion.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Local Ranking Problem on the BrowseGraph

Trevisiol

Aiello

Boldi

et al. 2015

Proceedings of the 38th International ACM SIGIR Conference on Research and Development in Information Retrieval

Self Cite

View full text Add to dashboard Cite

The "Local Ranking Problem" (LRP) is related to the computation of a centrality-like rank on a local graph, where the scores of the nodes could significantly differ from the ones computed on the global graph. Previous work has studied LRP on the hyperlink graph but never on the BrowseGraph, namely a graph where nodes are webpages and edges are browsing transitions. Recently, this graph has received more and more attention in many different tasks such as ranking, prediction and recommendation. However, a webserver has only the browsing traffic performed on its pages (local BrowseGraph) and, as a consequence, the local computation can lead to estimation errors, which hinders the increasing number of applications in the state of the art. Also, although the divergence between the local and global ranks has been measured, the possibility of estimating such divergence using only local knowledge has been mainly overlooked. These aspects are of great interest for online service providers who want to: (i) gauge their ability to correctly assess the importance of their resources only based on their local knowledge, and (ii) take into account real user browsing fluxes that better capture the actual user interest than the static hyperlink network. We study the LRP problem on a BrowseGraph from a large news provider, considering as subgraphs the aggregations of browsing traces of users coming from different domains. We show that the distance between rankings can be accurately predicted based only on structural information of the local graph, being able to achieve an average rank correlation as high as 0.8.

show abstract