Data Mining Practical Machine Learning Tools and Techniques

Matrix Factorization Techniques for Recommender Systems

The Object Management Group's (OMG) Model Driven Architecture (MDA) strategy envisages a world where models play a more direct role in software production, being amenable to manipulation and transformation by machine. Model Driven Engineering (MDE) is wider in scope than MDA. MDE combines process and analysis with architecture. This article sets out a framework for model driven engineering, which can be used as a point of reference for activity in this area. It proposes an organisation of the modelling 'space' and how to locate models in that space. It discusses different kinds of mappings between models. It explains why process and architecture are tightly connected. It discusses the importance and nature of tools. It identifies the need for defining families of languages and transformations, and for developing techniques for generating/configuring tools from such definitions. It concludes with a call to align metamodelling with formal language engineering techniques.

Model Driven Engineering

A recommender system aims to provide users with personalized online product or service recommendations to handle the increasing online information overload problem and improve customer relationship management. Various recommender system techniques have been proposed since the mid-1990s, and many sorts of recommender system software have been developed recently for a variety of applications. Researchers and managers recognize that recommender systems offer great opportunities and challenges for business, government, education, and other domains, with more recent successful developments of recommender systems for real-world applications becoming apparent. It is thus vital that a high quality, instructive review of current trends should be conducted, not only of the theoretical research results but more importantly of the practical developments in recommender systems. This paper therefore reviews up-to-date application developments of recommender systems, clusters their applications into eight main categories: e-government, e-business, e-commerce/e-shopping, e-library, e-learning, e-tourism, e-resource services and e-group activities, and summarizes the related recommendation techniques used in each category. It systematically examines the reported recommender systems through four dimensions: recommendation methods (such as CF), recommender systems software (such as BizSeeker), real-world application domains (such as e-business) and application platforms (such as mobile-based platforms). Some significant new topics are identified and listed as new directions. By providing a state-of-the-art knowledge, this survey will directly support researchers and practical professionals in their understanding of developments in recommender system applications. Research papers on various recommender system applications are summarized.The recommender systems are examined systematically through four dimensions.The recommender system applications are classified into eight categories.Related recommendation techniques in each category are identified.Several new recommendation techniques and application areas are uncovered.

Recommender system application developments

Recommender system is a specific type of intelligent systems, which exploits historical user ratings on items and/or auxiliary information to make recommendations on items to the users. It plays a critical role in a wide range of online shopping, e-commercial services and social networking applications. Collaborative filtering (CF) is the most popular approaches used for recommender systems, but it suffers from complete cold start (CCS) problem where no rating record are available and incomplete cold start (ICS) problem where only a small number of rating records are available for some new items or users in the system. In this paper, we propose two recommendation models to solve the CCS and ICS problems for new items, which are based on a framework of tightly coupled CF approach and deep learning neural network. A specific deep neural network SADE is used to extract the content features of the items. The state of the art CF model, timeSVD++, which models and utilizes temporal dynamics of user preferences and item features, is modified to take the content features into prediction of ratings for cold start items. Extensive experiments on a large Netflix rating dataset of movies are performed, which show that our proposed recommendation models largely outperform the baseline models for rating prediction of cold start items. The two proposed recommendation models are also evaluated and compared on ICS items, and a flexible scheme of model retraining and switching is proposed to deal with the transition of items from cold start to non-cold start status. The experiment results on Netflix movie recommendation show the tight coupling of CF approach and deep learning neural network is feasible and very effective for cold start item recommendation. The design is general and can be applied to many other recommender systems for online shopping and social networking applications. The solution of cold start item problem can largely improve user experience and trust of recommender systems, and effectively promote cold start items.

https://discovery.dundee.ac.uk/ws/files/39285097/Recommendation_system_for_cold_start_items.pdf

Collaborative filtering and deep learning based recommendation system for cold start items

Exploiting temporal context has been proved to be an effective approach to improve recommendation performance, as shown, e.g. in the Netflix Prize competition. Time-aware recommender systems (TARS) are indeed receiving increasing attention. A wide range of approaches dealing with the time dimension in user modeling and recommendation strategies have been proposed. In the literature, however, reported results and conclusions about how to incorporate and exploit time information within the recommendation processes seem to be contradictory in some cases. Aiming to clarify and address existing discrepancies, in this paper we present a comprehensive survey and analysis of the state of the art on TARS. The analysis show that meaningful divergences appear in the evaluation protocols used--metrics and methodologies. We identify a number of key conditions on offline evaluation of TARS, and based on these conditions, we provide a comprehensive classification of evaluation protocols for TARS. Moreover, we propose a methodological description framework aimed to make the evaluation process fair and reproducible. We also present an empirical study on the impact of different evaluation protocols on measuring relative performances of well-known TARS. The results obtained show that different uses of the above evaluation conditions yield to remarkably distinct performance and relative ranking values of the recommendation approaches. They reveal the need of clearly stating the evaluation conditions used to ensure comparability and reproducibility of reported results. From our analysis and experiments, we finally conclude with methodological issues a robust evaluation of TARS should take into consideration. Furthermore we provide a number of general guidelines to select proper conditions for evaluating particular TARS.

/pdf/time-aware-recommender-systems-a-comprehensive-survey-and-40qfzy0zif.pdf

Time-aware recommender systems: a comprehensive survey and analysis of existing evaluation protocols

In the Social Web, a number of diverse recommendation approaches have been proposed to exploit the user generated contents available in the Web, such as rating, tagging, and social networking information. In general, these approaches naturally require the availability of a wide amount of these user preferences. This may represent an important limitation for real applications, and may be somewhat unnoticed in studies focusing on overall precision, in which a failure to produce recommendations gets blurred when averaging the obtained results or, even worse, is just not accounted for, as users with no recommendations are typically excluded from the performance calculations. In this article, we propose a coverage metric that uncovers and compensates for the incompleteness of performance evaluations based only on precision. We use this metric together with precision metrics in an empirical comparison of several social, collaborative filtering, and hybrid recommenders. The obtained results show that a better balance between precision and coverage can be achieved by combining social-based filtering (high accuracy, low coverage) and collaborative filtering (low accuracy, high coverage) recommendation techniques. We thus explore several hybrid recommendation approaches to balance this trade-off. In particular, we compare, on the one hand, techniques integrating collaborative and social information into a single model, and on the other, linear combinations of recommenders. For the last approach, we also propose a novel strategy to dynamically adjust the weight of each recommender on a user-basis, utilizing graph measures as indicators of the target user's connectedness and relevance in a social network.

http://ir.ii.uam.es/reshet/pub/tist-camra11.pdf

An empirical comparison of social, collaborative filtering, and hybrid recommenders

In this paper, we describe the experiments conducted by the Information Retrieval Group at the Universidad Autonoma de Madrid (Spain) in order to better recommend movies for the 2010 CAMRa Challenge edition. Experiments were carried out on the dataset corresponding to weekly Filmtipset track. We consider simple strategies for taking into account the temporal context for movie recommendations, mainly based on variations of the KNN algorithm, which has been deeply studied in the literature, and one ad-hoc strategy, taking advantage of particular information in the weekly Filmtipset track. Results show that the usage of information near to the recommendation date alone can help improving recommendation results, with the additional benefit of reducing the information overload of the recommender engine. Furthermore, the use of social interaction information shows also a contribution in order to better predict a part of users' tastes.

http://ir.ii.uam.es/rim3/publications/camra10-campos.pdf

Simple time-biased KNN-based recommendations

Context-aware recommender systems have been proven to improve the performance of recommendations in a wide array of domains and applications. Despite individual improvements, little work has been done on comparing different approaches, in order to determine which of them outperform the others, and under what circumstances. In this paper we address this issue by conducting an empirical comparison of several pre-filtering, post-filtering and contextual modeling approaches on the movie recommendation domain. To acquire confident contextual information, we performed a user study where participants were asked to rate movies, stating the time and social companion with which they preferred to watch the rated movies. The results of our evaluation show that there is neither a clear superior contextualization approach nor an always best contextual signal, and that achieved improvements depend on the recommendation algorithm used together with each contextualization approach. Nonetheless, we conclude with a number of cues and advices about which particular combinations of contextualization approaches and recommendation algorithms could be better suited for the movie recommendation domain.

/pdf/context-aware-movie-recommendations-an-empirical-comparison-3g2frtker5.pdf

Context-Aware Movie Recommendations: An Empirical Comparison of Pre-filtering, Post-filtering and Contextual Modeling Approaches

The use of temporal dynamic terms in Matrix Factorization (MF) models of recommendation have been proposed as a means to obtain better accuracy in rating prediction task. However, the way such models have been tested may not be a realistic setting for recommendation. In this paper, we evaluated rating prediction and top-N recommendation tasks using a MF model with and without temporal dynamic terms under two evaluation settings. Our experiments show that the addition of dynamic parameters do not necessarily yield to better results on these tasks when a more strict time-aware separation of train/test data is performed, and moreover, results may vary notably when different evaluation schemes are used.

/pdf/towards-a-more-realistic-evaluation-testing-the-ability-to-3645o4egpv.pdf

Fernando Díez

Papers

Time-aware recommender systems: a comprehensive survey and analysis of existing evaluation protocols

An empirical comparison of social, collaborative filtering, and hybrid recommenders

Simple time-biased KNN-based recommendations

Context-Aware Movie Recommendations: An Empirical Comparison of Pre-filtering, Post-filtering and Contextual Modeling Approaches

Towards a more realistic evaluation: testing the ability to predict future tastes of matrix factorization-based recommenders