Session-aware recommendation: A surprising quest for the state-of-the-art

Latifi, Sara; Mauro, Noemi; Jannach, Dietmar

doi:10.1016/j.ins.2021.05.048

Cited by 42 publications

(32 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the authors solely describe these techniques and they do not perform an offline evaluation. [11], [13], [14] present evaluations of several recommendation techniques, including some intent-based ones against datasets of several domains. However, these works are not from the VoD domain, no user behavior analysis is performed, and their goal is to benchmark several recommendation techniques in the general case.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

From Data Analysis to Intent-Based Recommendation: An Industrial Case Study in the Video Domain

et al. 2022

View full text Add to dashboard Cite

This work presents a comprehensive study, from an industrial perspective, of the process between the collection of raw data, and the generation of next-item recommendation, in the domain of Video-on-Demand (VoD). Most research papers focus their efforts on analyzing recommender systems on already-processed datasets, but they do not face the same challenges that occur naturally in industry, e.g., processing raw interactions logs to create datasets for testing. This paper describes the whole process between data collection and recommendation, including cleaning, processing, feature engineering, session inferring, and all the challenges that a dataset provided by an industrial player in the domain posed. Then, a comparison on the new dataset of several intent-based recommendation techniques in the nextitem recommendation task follows, studying the impact of different factors like the session length, and the number of previous sessions available for a user. The results show that taking advantage of the sequential data available in the dataset benefits recommendation quality, since deep learning algorithms for sessionaware recommendation are consistently the most accurate recommenders. Lastly, a summary of the different challenges in the VoD domain is proposed, discussing on the best algorithmic solutions found, and proposing future research directions to be conducted based on the results obtained.

show abstract

Section: Related Workmentioning

confidence: 99%

“…In Section III-D, the results of the analysis are described. These results were then used to filter sessions as shown in Section III-E. (v) In order 1 Please, refer to [11]- [14] for surveys on intent-based recommendation.…”

Section: Introductionmentioning

confidence: 99%

From Data Analysis to Intent-Based Recommendation: An Industrial Case Study in the Video Domain

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Multi-session-based SBRSs incorporate other sessions to complement the information in the current session for next-item recommendations in it. Some multi-session-based SBRSs, e.g., session-aware recommender systems [5,21], incorporate the historical sessions of the current user. For example, both hierarchical RNN (HRNN) [14] and inter-and intra-session RNNs (II-RNN) [15] first employ a session-level RNN and an item-level RNN to encode a sequence of historical sessions of the current user and a sequence of items in the current session, respectively, and then combines the outputs from both RNNs to predict the next item in the current session.…”

Section: Related Workmentioning

confidence: 99%

“…(5) 4.2 Global ModuleTo effectively learn useful prior knowledge from other sessions, we first form two candidate similar session sets as the inputs of the global module. The first is the set of 𝑢 𝑐 's previous sessions H(𝑢 𝑐 ) and the second set S(𝑢 𝑐 ) contains the sessions of a few other users who have similar preferences with 𝑢 𝑐 .…”

mentioning

confidence: 99%

Next-item Recommendations in Short Sessions

Song

Wang

et al. 2021

Fifteenth ACM Conference on Recommender Systems

View full text Add to dashboard Cite

The changing preferences of users towards items trigger the emergence of session-based recommender systems (SBRSs), which aim to model the dynamic preferences of users for next-item recommendations. However, most of the existing studies on SBRSs are based on long sessions only for recommendations, ignoring short sessions, though short sessions, in fact, account for a large proportion in most of the real-world datasets. As a result, the applicability of existing SBRSs solutions is greatly reduced. In a short session, quite limited contextual information is available, making the next-item recommendation very challenging. To this end, in this paper, inspired by the success of few-shot learning (FSL) in effectively learning a model with limited instances, we formulate the next-item recommendation as an FSL problem. Accordingly, following the basic idea of a representative approach for FSL, i.e., meta-learning, we devise an effective SBRS called INter-SEssion collaborative Recommender neTwork (INSERT) for next-item recommendations in short sessions. With the carefully devised local module and global module, INSERT is able to learn an optimal preference representation of the current user in a given short session. In particular, in the global module, a similar session retrieval network (SSRN) is designed to find out the sessions similar to the current short session from the historical sessions of both the current user and other users, respectively. The obtained similar sessions are then utilized to complement and optimize the preference representation learned from the current short session by the local module for more accurate next-item recommendations in this short session. Extensive experiments conducted on two real-world datasets demonstrate the superiority of our proposed INSERT over the state-of-the-art SBRSs when making next-item recommendations in short sessions.

show abstract

“…Through their studies, they found that much of the reported progress only seems to be "virtual", as the latest models are almost always outperformed by existing methods (see also Rendle et al [28] for a related analysis). Various reasons may contribute to this surprising phenomenon, including the choice of (too weak) baselines [18,21] or the lack of a proper tuning of the baselines. Moreover, in such independent evaluations, i.e., that are not done by authors of the compared methods, it often turns out that there is no clear winner across datasets and accuracy measures.…”

Section: Introductionmentioning

confidence: 99%

Top-N Recommendation Algorithms: A Quest for the State-of-the-Art

Anelli,

Bellogín,

Di Noia

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Research on recommender systems algorithms, like other areas of applied machine learning, is largely dominated by efforts to improve the state-of-the-art, typically in terms of accuracy measures. Several recent research works however indicate that the reported improvements over the years sometimes "don't add up", and that methods that were published several years ago often outperform the latest models when evaluated independently. Different factors contribute to this phenomenon, including that some researchers probably often only fine-tune their own models but not the baselines.In this paper, we report the outcomes of an in-depth, systematic, and reproducible comparison of ten collaborative filtering algorithmscovering both traditional and neural models-on several common performance measures on three datasets which are frequently used for evaluation in the recent literature. Our results show that there is no consistent winner across datasets and metrics for the examined top-n recommendation task. Moreover, we find that for none of the accuracy measurements any of the considered neural models led to the best performance. Regarding the performance ranking of algorithms across the measurements, we found that linear models, nearestneighbor methods, and traditional matrix factorization consistently perform well for the evaluated modest-sized, but commonly-used datasets. Our work shall therefore serve as a guideline for researchers regarding existing baselines to consider in future performance comparisons. Moreover, by providing a set of fine-tuned baseline models for different datasets, we hope that our work helps to establish a common understanding of the state-of-the-art for top-n recommendation tasks.

show abstract

Session-aware recommendation: A surprising quest for the state-of-the-art

Cited by 42 publications

References 16 publications

From Data Analysis to Intent-Based Recommendation: An Industrial Case Study in the Video Domain

From Data Analysis to Intent-Based Recommendation: An Industrial Case Study in the Video Domain

Next-item Recommendations in Short Sessions

Top-N Recommendation Algorithms: A Quest for the State-of-the-Art

Contact Info

Product

Resources

About