Efficient algorithms for processing preference queries

Recommender systems play an important role in most modern e-commerce applications. They have allowed users to become aware of the myriad choices available to them. The ease of information and the abundance of options have helped users make educated decisions. A recommender system studies a user's preferences and continues learning the user's changing interests, so as to suggest items that incline with the user's interests. In cases where a user is new to the application, or the user prefers not to discourse preferences, the recommender system is unable to gather the user's preference on any item. This is called the cold start problem; wherein the system can make valid recommendations only once the user starts informing the system about his/her choices. In this paper, we discuss the challenges faced by the cold start problem and how this problem may be alleviated using social media. We suggest an approach where we collect public information from users' social media accounts and analyze this information to understand their preferences. In particular, we gather the new user's information using their Twitter profile; i.e., the user's interest and preferences are extracted from his/her Twitter profile by analyzing his/her tweets. These interests will help the system understand what kind of movies the user will be most interested in. We compare these preferences with the metadata about the individual items. Using this approach, we develop a movie recommendation system wherein we produce top-N movie recommendations for a user. We used the MovieTweetings dataset to model the application. Two sets of results have been produced. In the first, smaller set of 770 users, 72.67% of users have received 100% accurate movie recommendations while nearly 80% of users got more than 75% accuracy. For the second, larger set of more than 3,500 users, 53 % of users have received 100% accurate recommendations while 72% of users got more than 75% accuracy. These encouraging results have demonstrated that the approach is effectively in alleviating cold start problems in recommendation systems, and may be applicable to many other e-commerce applications.

Today, advertisement has become a part of media content and it is exposed to viewers in ways such as home shopping, PPL (Product Placement), and T-commerce. Even though media contents and platforms have been developed into user-interactive services, the interaction between advertisement systems and users is still limited. For advertising services to interact with users, a recommendation system should be introduced. Recommendation systems have been studied such as content-based, collaborative filtering, and hybrid methods. Since conventional recommendation systems require high computational complexity in the preference prediction process and do not consider the time factor, there are limitations in performing recommendation considering user preference. In this paper, we propose an advertisement recommendation system that predicts the ATU’s (Active Target User) preference for items in real-time. The proposed system creates PRETREE by modeling ATU's history data with a tree data structure for reducing computational complexity. A recommendation is triggered when the item that the ATU is looking at in the media content is detected. If there are rating records for the item in the history of the ATU, the preferences are predicted on the basis of PRETREE. If there is no history for the item, the proposed system predicts preference on the basis of similar groups and the LCC (Longest Common Category) node at the ATU's PRETREE. Finally, we evaluate the performance of the proposed recommendation system using real- world data.

Typical tag recommendation systems for photos shared on social networks such as Flickr, use visual content analysis, collaborative filtering or personalization strategies to produce annotations. However, the dependence on manual intervention and the knowledge of sufficient personal preferences coupled with the folksonomic issues limit the scope of these strategies. In this paper, we present a fully automatic and folksonomically scalable tag recommendation model that can recommend tags for a user's photos without an explicit knowledge of the user's personal tagging preferences. The model is learned using the collective tagging behavior of other users in the user's local interaction network, which we believe approximates the user's preferences, at least partially. The tag recommendation model generates content-based annotations and then uses a Naive Bayes formulation to translate these annotations to a set of folksonomic tags selected from the tags used by the users in the local interaction network. Quantitative and qualitative comparisons with 890 Flickr networks show that this approach is highly useful for tag recommendation in the presence of insufficient information of a user's own preferences.

A location-based orientation-aware recommender system using IoT smart devices and Social Networks

Recommender systems usually suggest items by exploiting all the previous interactions of the users with a system (e.g., in order to decide the movies to recommend to a user, all the movies she previously purchased are considered). This canonical approach sometimes could lead to wrong results due to several factors, such as a change in user preferences over time, or the use of her account by third parties. This kind of incoherence in the user profiles defines a lower bound on the error the recommender systems may achieve when they generate suggestions for a user, an aspect known in literature as magic barrier. This paper proposes a novel dynamic coherence-based approach to define the user profile used in the recommendation process. The main aim is to identify and remove, from the previously evaluated items, those not semantically adherent to the others, in order to make a user profile as close as possible to the user's real preferences, solving the aforementioned problems. Moreover, reshaping the user profile in such a way leads to great advantages in terms of computational complexity, since the number of items considered during the recommendation process is highly reduced. The performed experiments show the effectiveness of our approach to remove the incoherent items from a user profile, increasing the recommendation accuracy.

Presenting small chunks of interesting information to a target user from the large pool of information available on World Wide Web are the primary task of a recommender system. Memory based Collaborative Filtering generates recommendations based on preferences of those users whose past preferences are similar to the current preferences of the target user. These users collectively form the Nearest Neighbor Set (NN Set) of the target user. The better the selection of NN Set, the better is the generation of recommendations. In the proposed scheme, the user ratings (preferences) on the available items are divided into two levels. Level I consist of ratings on popular items and Level II consist of ratings on unpopular items. The proposed similarity measure, “TSimD(U <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">X1</sub> ,U <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">X2</sub> )”, between two users is based on weighted difference entropy. Modified memory based Collaborative Filtering calculates the proposed similarity measure at both the levels to improve the selection of users in the NN Set of the target user. It selects those users in the NN Set who are more similar to the target user with respect to items at Level II as compared to similarity between them with respect to items at Level I. The results on Movie Lens dataset depicted that the proposed similarity measure had higher accuracy than the Weighted Difference Entropy based similarity measure which worked on the entire preferences of the users. The proposed similarity measure was also compared with other similarity measures (like Cosine, Pearson Correlation, Spearman and Rating Frequency based Similarity Measure) which were also obtained at two levels.

Recent advances in mobile devices and their sensing capabilities have enabled the collection of rich contextual information and mobile device usage records through the device logs. These context-rich logs open a venue for mining the personal preferences of mobile users under varying contexts and thus enabling the development of personalized context-aware recommendation and other related services, such as mobile online advertising. In this article, we illustrate how to extract personal context-aware preferences from the context-rich device logs, or context logs for short, and exploit these identified preferences for building personalized context-aware recommender systems. A critical challenge along this line is that the context log of each individual user may not contain sufficient data for mining his or her context-aware preferences. Therefore, we propose to first learn common context-aware preferences from the context logs of many users. Then, the preference of each user can be represented as a distribution of these common context-aware preferences. Specifically, we develop two approaches for mining common context-aware preferences based on two different assumptions, namely, context-independent and context-dependent assumptions, which can fit into different application scenarios. Finally, extensive experiments on a real-world dataset show that both approaches are effective and outperform baselines with respect to mining personal context-aware preferences for mobile users.

In mathematics and statistics, the desire to eliminate mathematical tedium and facilitate exploration has lead to computer algebra systems. These computer algebra systems allow students and researchers to perform more of their work at a conceptual level. The design of generic algorithms for tedious computations allows modelers to push current modeling boundaries outward more quickly. Probability theory, with its many theorems and symbolic manipulations of random variables is a discipline in which automation of certain processes is highly practical, functional, and efficient. There are many existing statistical software packages, such as SPSS, SAS, and S-Plus, that have numeric tools for statistical applications. There is a potential for a probability package analogous to these statistical packages for manipulation of random variables. The software package being developed as part of this dissertation, referred to as “A Probability Programming Language” (APPL) is a random variable manipulator and is proposed to fill a technology gap that exists in probability theory. My research involves developing algorithms for the manipulation of discrete random variables. By defining data structures for random variables and writing algorithms for implementing common operations, more interesting and mathematically intractable probability problems can be solved, including those not attempted in undergraduate statistics courses because they were deemed too mechanically arduous. Algorithms for calculating the probability density function of order statistics, transformations, convolutions, products, and minimums/maximums of independent discrete random variables are included in this dissertation.

In recommender systems user preferences can be fairly dynamic, as users tend to exploit a wide range of items and modify their tastes accordingly over time. In this paper, we model user-item interactions over time using a tensor that has time as a dimension (mode). To account for the fact that user preferences change individually, we propose a new measure of user-preference dynamics (UPD) that captures the rate with which the current preferences of each user have been shifted. UPD shows the variability in how users interact with items in recommender systems. We generate recommendations based on a tensor factorization technique, where the importance of past user preferences are weighted according to their UPD values, that is, higher UPD values downweigh more past user preferences. Additionally, we exploit users’ side data, such as demographics, which improve the accuracy of recommendations based on a coupled tensor-matrix factorization scheme. Our empirical evaluation uses two real benchmark datasets from the social media platforms Last.fm and MovieLens, containing users’ history records pertaining to listening to songs and viewing movies, respectively. We demonstrate that in both datasets, there are users with a varying level of dynamics, expressed by the UPD metric. Our experimental results show that the proposed method outperforms several baselines, by taking into account both dynamics and side data of users.

This position paper discusses a novel recommender system for e-commerce in virtual reality environments. The system provides recommendations by taking into account prepurchase ratings in addition to traditional postpurchase ratings. Users' positive emotions are captured in the form of electroencephalogram (EEG) signals while interacting with 3D virtual products prior to purchase. The prepurchase ratings are calculated from the averaged relative power of the collected EEG signals. Prepurchase ratings are complementary to postpurchase ratings and help in alleviating two severe issues that traditional recommender systems suffer from: data sparsity and cold start. By making proper use of both pre- and postpurchase ratings, user preference can be modeled more accurately. This will improve the effectiveness of the current recommender systems and may change the traditional e- business applications. 

Invariant user preference learning is a core task in recommender systems. Accurately modeling user preferences is crucial, as it directly impacts the quality of recommendations. However, heterogeneous user preferences in feedback data often exhibit mixture distributions, which obscure invariant preferences and introduce bias. Existing methods typically address this issue by partitioning feedback data into multiple environments and learning invariant preferences across them. Nonetheless, these approaches often lack theoretical guarantees for environment construction and fail to capture the dynamic nature of user preferences across different environments. Along these lines, we propose a novel framework, Invariant and Environment-specific Preferences for unbiased recommendation (IEPref). IEPref leverages auxiliary information as a reliable signal to guide the environment classifier in partitioning the environment, thereby enabling the learning of more stable and generalizable user preference with theoretical guarantees. Additionally, we design environment-specific proxy modules to capture context-dependent preference patterns unique to each environment. The environment classifier assigns each user-item interaction to its corresponding latent environment, and both invariant and environment-specific preferences are integrated for recommendations. Extensive experiments on five real-world datasets demonstrate that IEPref achieves superior performance over existing baselines, effectively mitigating recommendation bias while preserving personalized modeling capabilities.

Interactive evolutionary algorithms have been used successfully for search and optimization problems. However, user evaluation fatigue and uncertainties greatly restrict their application in complex optimization scenarios. Implicit evaluation techniques can reduce user fatigue effectively, but methods for handling evaluation uncertainties have not yet been explored thoroughly. We study an interactive evolutionary algorithm with uncertain preferences using an implicit evaluation model. First, we design a Gaussian function to define evaluation uncertainties based on interactions. Then, by combining the evaluation uncertainties and the interactions, we calculate the preference weights for each variable, with which a weighted Conditional Preference Net is constructed to provide quantitative depiction of the user's evaluation preference. Using the weighted Net, we estimate the fitness of an individual for the achievement of the evolutionary process. In the iteration, the Net is updated according to newly performed interactions to ensure reliable tracking of the user's preference. We apply a genetic algorithm-based interactive evolutionary algorithm to a personalized search for books. The proposed algorithm accelerates search speed, alleviates user fatigue, and achieves higher rates of satisfying solutions.

ExUP recommendations: Inferring user's product metadata preferences from single-criterion rating systems

Recommender systems have been widely used by online service providers to conduct targeted marketing. However, since user behaviour and preferences are fluid and dynamic, predicting users’ online actions is indeed a very difficult task. Variational Autoencoders (VAEs) have been recently proposed to improve the prediction accuracy of user preferences. Although classic VAEs go beyond linear modeling, they suffer from underfitting problem when the underlying datasets are sparse. Moreover, we found that classic VAEs are ineffective for dynamic user preferences. To address these deficiencies, we propose Short Term Memory Variational Autoencoder (STMVAE) to overcome the underfitting issue and to better handle the dynamics. This is achieved by capturing users’ short term preferences to generate near term predictions. The validity and efficacy of our proposed model are evaluated comprehensively using three datasets in our experiments.

The vast majority of recommender systems model preferences as static or slowly changing due to observable user experience. However, spontaneous changes in user preferences are ubiquitous in many domains like media consumption and key factors that drive changes in preferences are not directly observable. These latent sources of preference change pose new challenges. When systems do not track and adapt to users' tastes, users lose confidence and trust, increasing the risk of user churn. We meet these challenges by developing a model of novelty preferences that learns and tracks latent user tastes. We combine three innovations: a new measure of item similarity based on patterns of consumption co-occurrence; model for spontaneous changes in preferences; and a learning agent that tracks each user's dynamic preferences and learns individualized policies for variety. The resulting framework adaptively provides users with novelty tailored to their preferences for change per se.