Browsing by Subject "Recommender Systems"

Now showing 1 - 6 of 6
  • Thumbnail Image
    Item
    Computational techniques for more accurate and diverse recommendations.
    (2011-08) Kwon, YoungOk
    Recommender systems are becoming an increasingly important research area due to the growing demand for personalized recommendations. The volume of information available to each user and the number of products carried in e-commerce marketplaces have grown tremendously. Thus, recommender systems are needed to help individual users find the most relevant items from an enormous number of choices and eventually increase sales by exposing users to what they may like, but may not have considered otherwise. Despite significant progress in developing new recommendation techniques within both industry and academia, most research, to date, has focused on improving recommendation accuracy (i.e., the accuracy with which the recommender system predicts users` ratings for items they have not yet rated). While recommendation accuracy is undoubtedly important, there is a growing understanding that accuracy does not always imply usefulness to users. Therefore, in addition to investigating the accuracy of recommendations, my dissertation also considers the diversity of recommendations as another important aspect of recommendation quality and explores the relationship between accuracy and diversity. The diversity of recommendations can be expressed by the number of unique items recommended across all users, which reflects the ability of recommender systems to go beyond the obvious, best-selling items, and to generate more idiosyncratic, personalized, and long-tail recommendations. This dissertation presents four studies which propose new recommendation approaches that can improve accuracy and diversity. The first study enhances traditional recommendation algorithms by augmenting them with multi-criteria rating information for more accurate recommendations. The second study applies heuristic-based ranking approaches for more diverse recommendations. The third study develops more sophisticated optimization approaches for direct diversity maximization. The fourth study explores the possible combinations of the two types of approaches - incorporation of multi-criteria rating information and the use of different ranking methods - as a way to generate recommendations that are both more accurate and more diverse. The new recommendation approaches proposed in this dissertation enrich the body of knowledge on recommender systems by extending single-rating recommendation problems to address multi-criteria recommendation problems and exploring new ways to tackle the accuracy-diversity tradeoff issue. Individual users and online content providers will also benefit from the proposed approaches, in that each user will find more relevant and personalized items from more accurate and diverse recommendations provided by recommender systems. These approaches could potentially lead to increased loyalty and sales, thus, benefiting the providers as well.
  • Thumbnail Image
    Item
    Deep reinforcement learning for personalized treatment recommendation
    (2023-03) Liu, Mingyang
    In precision medicine, the ultimate goal is to recommend the most effective treatment to an individual patient based on patient-specific molecular and clinical profiles, possibly high-dimensional. To advance cancer treatment, large-scale screenings of cancer cell lines against chemical compounds have been performed to help better understand the relationship between genomic features and drug response; existing machine learning approaches use exclusively supervised learning, including penalized regression and recommender systems. When there is only one time point, it refers to individualized treatment selection, which is employed to maximize a certain clinical outcome of a specific patient based on a patient's clinical or genomic characteristics, given a patients' heterogeneous response to treatments. Although developing such a rule is conceptually important to personalized medicine, existing methods such as the $L_1$-penalized least squares \citep{qian2011performance} suffers from the difficulty of indirect maximization of clinical outcome, while the outcome weighted learning \citep{zhao2012estimating} directly maximizing the clinical outcome is not robust against any perturbation of the outcome. We will first propose a weighted $\psi$-learning method to optimize an individualized treatment rule, which is robust again perturbation of data near decision boundary through the notation of separation. To deal with nonconvex minimization, we employ a difference of convex algorithm to solve the non-convex minimization iteratively based on a decomposition of the cost function into a difference of two convex function. On this ground, we also introduce a variable selection method for further removing redundant variables for higher performance. Finally, we illustrate the proposed method through simulations and a lung health study, and demonstrate that it yields higher performance in terms of accuracy of prediction of individualized treatment. However, it would be more efficient to apply reinforcement learning (RL) to sequentially learn as data accrue, including selecting the most promising therapy for a patient given individual molecular and clinical features and then collecting and learning from the corresponding data. In this way, we propose a novel personalized ranking system called Proximal Policy Optimization Ranking (PPORank), which ranks the drugs based on their predicted effects per cell line (or patient) in the framework of deep reinforcement learning (DRL). Modeled as a Markov decision process (MDP), the proposed method learns to recommend the most suitable drugs sequentially and continuously over time. As a proof-of-concept, we conduct experiments on two large-scale cancer cell line data sets in addition to simulated data. The results demonstrate that the proposed DRL-based PPORank outperforms the state-of-the-art competitors based on supervised learning. Taken together, we conclude that novel methods in the framework of DRL have great potential for precision medicine and should be further studied.
  • Thumbnail Image
    Item
    Improvements in Holistic Recommender System Research
    (2018-08) Kluver, Daniel
    Since the mid 1990s, recommender systems have grown to be a major area of deployment in industry, and research in academia. A through-line in this research has been the pursuit, above all else, of the perfect algorithm. With this admirable focus has come a neglect of the full scope of building, maintaining, and improving recommender systems. In this work I outline a system deployment and a series of offline and online experiments dedicated to improving our holistic understanding of recommender systems. This work explores the design, algorithms, early performance, and interfaces of recommender systems within the scope of how they are interconnected with other aspects of the system. This work explores many indivisual aspects of a recommender system while keeping in mind how they are connected to other aspects of the system. The contributions of this thesis are: an exploration of the design of the BookLens system, a prototype recommender system for library-item recommendation; a methodology and exploration of algorithm performance for users with very few ratings which shows that the popular Item-Item recommendation algorithm performs very poorly in this context; an exploration of the issues faced by Item-Item, as well as fixes for these issues confirmed by both an offline and online analysis; and finally, the preference bits model for measuring the amount of noise and information contained in user ratings, as well as a rating support interface capable of reducing the noise in user ratings leading to superior algorithm performance. Supporting these contributions are the following specific methodological improvements: a bias free methodology for measuring algorithm performance over a range of profile sizes; a prototype user-study design for investigating new-user recommendation through Amazon Mechanical Turk; the preference bits model as well as derived measurements of preference bits per rating, per impressions, and per second; and finally a sound experimental design that can be used to empirically measure preference bits values for a given interface. It is our hope that these methodological contributions can help researchers in the recommender systems field ask new questions and further the holistic study of recommender systems.
  • Thumbnail Image
    Item
    Improving the Quality of Top-N Recommendation
    (2018-02) Christakopoulou, Evangelia
    Top-N recommenders are systems that provide a ranked list of N products to every user; the recommendations are of items that the user will potentially like. Top-N recommendation systems are present everywhere and used by millions of users, as they enable them to quickly find items they are interested in, without having to browse or search through big datasets; an often impossible task. The quality of the recommendations is crucial, as it determines the usefulness of the recommender to the users. So, how do we decide which products should be recommended? Also, how do we address the limitations of current approaches, in order to achieve better quality? In order to provide insight into these problems, this thesis focuses on developing novel, scalable algorithms that improve the state-of-the-art top-N recommendation quality, while providing insight into the top-N recommendation task. The developed algorithms address some of the limitations of existent top-N recommendation approaches and can be applied to real-world problems and datasets. The main areas of our contributions are the following: 1. Exploiting higher-order sets of items: We investigate to what extent higher-order sets of items are present in real-world datasets, beyond pairs of items. We also show how to best utilize them to improve the top-N recommendation quality. 2. Estimating a global and multiple local models: We show that estimating multiple user-subset specific local models, beyond a global model significantly improves the top-N recommendation quality. We demonstrate this with both item-item models and latent space models. 3. Investigating and using the error: We investigate what are the properties of the error and how they correlate with the top-N recommendation quality, in methods that treat the missing entries as zeros. Then, we utilize the learned insights to develop a method, which explicitly uses the error. We have applied our algorithms to big datasets, with millions of ratings, that span different areas, such as grocery transactions, movie ratings, and retail transactions, showing significant improvements over the state-of-the-art.
  • Thumbnail Image
    Item
    Nurturing tagging communities
    (2009-03) Sen, Shilad Wieland
    Member contributions power many online communities. Users have uploaded billions of images to flickr, bookmarked millions pages on del.icio.us, and authored millions of encyclopedia articles at Wikipedia. Tags --- member contributed words or phrases that describe items --- have emerged as a powerful method for searching, organizing, and making sense of, these vast corpora. In this thesis we explore the dynamics, challenges, and possibilities of tagging systems. We study the way in which factors influencing an individual user's choice of tags can affect the evolution of community tags as a whole. Like other community-maintained systems, tagging systems can suffer from low quality contributions. We study interfaces and algorithms that can differentiate between low quality and high quality tags. Finally, we explore tagommenders, tag-based recommendation algorithms that combine the flexibility of tags with the automation of recommender systems. We base our explorations on tagging activity in the MovieLens movie recommendation system. We analyze tagging behavior, user studies, and surveys, of 97,000 tags and 3,600 users. Our results provide insight into the dynamics of existing tagging communities, and suggest mechanisms that address challenges of, and provide extensions to, tagging systems.
  • Thumbnail Image
    Item
    What is the Value of Rating Obscure Items? An Analysis of the Effect of Less-Popular Items on Recommendation Quality
    (2019-06) Narayan, Anshuman
    Recommender systems designers believe that the system stands to benefit from the users rating items that do not have many ratings. However, the effect of this act of rating lesser known items on the user’s recommendations is unknown. This leads to asking the question of whether these low popularity items affect the recommendations received by users. This work looks at the effect less popular items have on a user’s recommendations and the prediction and recommendations metrics that quantify the quality of recommendations. Using a matrix factorization model to build a recommender system, we modify a subset of users’ ratings data and look at the difference in recommendations generated. We also make use of popular recommender systems metrics such as nDCG, Precison and Recall to evaluate the effect of these modifications.Apart from looking at the ef- fect of this ”truncation” of casual user ratings data on the casual users themselves, we also look at the effects of this ”truncation” on the more invested users of the system, in terms of top-n recommendation and prediction metrics. The results of these evalu- ations appear promising, with very little to no loss of information, personalization or metric scores for more casual users. The results of these evaluations for more serious users also appears to have little effect on the performance of top-n recommendation and prediction metrics.