Browsing by Author "Chaitanya, V.S."

Now showing 1 - 4 of 4

A Clustering-based model for the Generation of Diversified Recommendations
(Institute of Electrical and Electronics Engineers Inc., 2022) Chaitanya, V.S.; Mohan, M.; Santhi Thilagam, P.S.
The primary goal of a recommender system is to generate accurate recommendations according to the user's interests. But the user's satisfaction increases when they get a chance to view the diverse categories of items. There exist several works on the generation of diverse recommendations but the performance of these methods often gets limited due to the issues such as cold start, filter bubble long tail, and grey sheep. Moreover, these methods do not consider the user's preference regarding exploration and exploitation while generating the recommendations. To this extent, this work proposes a model known as the iterative clustering-based diversity model, which can generate diverse recommendations and also solve the above-said issues. It groups the items based on the item description using the TF-IDF algorithm. The model generates two recommendations in such a way that one recommendation is similar and the other is different in comparison with the last interaction made by the user. The model has been evaluated on the benchmark dataset and has achieved promising results. Â© 2022 IEEE.
PRIDES: A Probabilistic Model for Recurrent User Interest Drift Identification in Session-Based Recommendation
(Institute of Electrical and Electronics Engineers Inc., 2025) Chaitanya, V.S.; Santhi Thilagam, P.S.
A Session-Based Recommendation (SBR) identifies correlations among session interactions to understand user preferences and generate appropriate recommendations. A key challenge in this context is the dynamic change in user preferences, particularly when preferences disappear and reappear within a session, a phenomenon referred to as Recurrent User Interest Drift (RUID). Effectively capturing RUID is significant for aligning recommendations with ongoing user preferences. Existing SBR approaches often misclassify user preferences that differ from other session interactions as noise (unintentional interactions), relying on dwell time (the amount of time a user spends viewing an item) or neighboring sessions, thereby overlooking their potential reappearance as RUID later in the session. To the best of our knowledge, this work is the first to address the challenge of identifying RUID in SBR. The proposed approach assigns probabilistic scores to each interaction by considering its similarity to the immediate previous interaction, its inclusion among popular items (items with a higher number of interactions), its similarity to previous interactions, and the dwell time. As user preference may reappear anytime during the session, and RUID identification requires analyzing subsequent interactions, a list-based approach is used to retain these interactions until the session ends, enabling effective RUID identification. The matrix factorization-based attentive session encoder incorporates both short-term (ongoing) preferences and long-term (historical) preferences to generate personalized recommendations. Experimental results on three benchmark datasets, Yoochoose, Last.fm, and Gowalla, show that our method outperforms 14 state-of-the-art baselines, achieving an improvement of 2.28% in recall@20 and 1.39% in Mean Reciprocal Rank (MRR@20) on Yoochoose, 3.58% in recall@20 and 2.70% in MRR@20 on Last.fm, and 5.35% in recall@20 and 4.17% in MRR@20 on Gowalla datasets. © 2013 IEEE.
Understanding dynamic user preferences using user-centric reservoir model for enhancing personalization
(Springer Science and Business Media B.V., 2025) Chaitanya, V.S.; Jain, H.; Santhi Thilagam, P.S.
Streaming Session-based Recommender Systems (SSRS) generate personalized recommendations from continuously arriving session data in the form of clickstreams. However, the massive arrival of clickstreams causes the forgetting of learned user preferences, necessitating periodic retraining of the SSRS model. This requires storing clickstreams in the reservoir; however, its limited storage space causes existing approaches to randomly replace and sample clickstreams, leading to the loss of user preferences that are unique compared to other users. Moreover, retraining the SSRS model with all the clickstreams in the reservoir results in an overload scenario. This work introduces a user-based partitioning approach that allocates dedicated reservoir partitions for individual user interactions. Furthermore, a priority-based sampling approach is introduced to select clickstreams from each partition for retraining. By considering both historical and ongoing user preferences, the proposed methodology generates personalized recommendations for the users. Experimental results on benchmark datasets demonstrate its superiority over state-of-the-art baseline methodologies. © Bharati Vidyapeeth's Institute of Computer Applications and Management 2025.
User Interest Drift Identification Using Contextual Factors inÂ Implicit Feedback-Based Recommender Systems
(Springer Science and Business Media Deutschland GmbH, 2023) Chaitanya, V.S.; Deo, S.; Santhi Thilagam, P.S.
The modeling of appropriate recommendations using the session interactions in the implicit feedback-based recommender systems necessitates the identification of user interest drift. But this identification is challenging due to the presence of unintentional interactions (noise) made by the user. Most of the existing literature focused on understanding the correlation between ongoing session interactions but did not explore the contextual factors, such as the time of occurrence of the session and the itemâ€™s popularity, that led the user to perform that specific interaction. This has resulted in the wrongful categorization of interactions between user interest drift and noise. To overcome these limitations, this work proposes a deep learning-based approach that uses both ongoing session information and contextual information. Depending on availability, this work also considers the userâ€™s previous interactions to generate personalized recommendations. In comparison with the existing works, this work effectively identifies the user interest drift and generates the appropriate recommendations for the users. The proposed approach demonstrates superior performance over state-of-the-art baselines in terms of Recall and MRR, as evidenced by experimental results on benchmark datasets. Â© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.