Recommender systems have developed in parallel with the web. They were initially based on demographic, content-based and collaborative filtering. Currently, these systems are incorporating social information. In the future, they will use implicit, local and personal information from the Internet of things. This article provides an overview of recommender systems as well as collaborative filtering methods and algorithms; it also explains their evolution, provides an original classification for these systems, identifies areas of future implementation and develops certain areas selected for past, present or future importance.

http://romisatriawahono.net/lecture/rm/survey/information%20retrieval/Bobadilla%20-%20Recommender%20Systems%20-%202013.pdf

Recommender systems survey

Concept drift primarily refers to an online supervised learning scenario when the relation between the input data and the target variable changes over time. Assuming a general knowledge of supervised learning in this article, we characterize adaptive learning processes; categorize existing strategies for handling concept drift; overview the most representative, distinct, and popular techniques and algorithms; discuss evaluation methodology of adaptive algorithms; and present a set of illustrative applications. The survey covers the different facets of concept drift in an integrated way to reflect on the existing scattered state of the art. Thus, it aims at providing a comprehensive introduction to the concept drift adaptation for researchers, industry analysts, and practitioners.

/pdf/a-survey-on-concept-drift-adaptation-xv7b4z5tkr.pdf

A survey on concept drift adaptation

Matrix Factorization Techniques for Recommender Systems

Collaborative filtering (CF) is a successful approach commonly used by many recommender systems. Conventional CF-based methods use the ratings given to items by users as the sole source of information for learning to make recommendation. However, the ratings are often very sparse in many applications, causing CF-based methods to degrade significantly in their recommendation performance. To address this sparsity problem, auxiliary information such as item content information may be utilized. Collaborative topic regression (CTR) is an appealing recent method taking this approach which tightly couples the two components that learn from two different sources of information. Nevertheless, the latent representation learned by CTR may not be very effective when the auxiliary information is very sparse. To address this problem, we generalize recently advances in deep learning from i.i.d. input to non-i.i.d. (CF-based) input and propose in this paper a hierarchical Bayesian model called collaborative deep learning (CDL), which jointly performs deep representation learning for the content information and collaborative filtering for the ratings (feedback) matrix. Extensive experiments on three real-world datasets from different domains show that CDL can significantly advance the state of the art.

/pdf/collaborative-deep-learning-for-recommender-systems-59tevxd83s.pdf

Collaborative Deep Learning for Recommender Systems

Among different recommendation techniques, collaborative filtering usually suffer from limited performance due to the sparsity of user-item interactions. To address the issues, auxiliary information is usually used to boost the performance. Due to the rapid collection of information on the web, the knowledge base provides heterogeneous information including both structured and unstructured data with different semantics, which can be consumed by various applications. In this paper, we investigate how to leverage the heterogeneous information in a knowledge base to improve the quality of recommender systems. First, by exploiting the knowledge base, we design three components to extract items' semantic representations from structural content, textual content and visual content, respectively. To be specific, we adopt a heterogeneous network embedding method, termed as TransR, to extract items' structural representations by considering the heterogeneity of both nodes and relationships. We apply stacked denoising auto-encoders and stacked convolutional auto-encoders, which are two types of deep learning based embedding techniques, to extract items' textual representations and visual representations, respectively. Finally, we propose our final integrated framework, which is termed as Collaborative Knowledge Base Embedding (CKE), to jointly learn the latent representations in collaborative filtering as well as items' semantic representations from the knowledge base. To evaluate the performance of each embedding component as well as the whole system, we conduct extensive experiments with two real-world datasets from different scenarios. The results reveal that our approaches outperform several widely adopted state-of-the-art recommendation methods.

Collaborative Knowledge Base Embedding for Recommender Systems

Formant Dynamics is an interesting research field in Speech Perception, Speech Parameterizacion, Synthesis, and Recognition. The Spectrogram shown in Fig. 1.a, corresponding to the utterance “I wished you were here a year ago” reveals important dynamic changes in the positions of the poles in the Transfer Function of the Vocal Tract with time. In the associated template shown in Fig. 1.b, the set of PARCOR parameters generating the Spectrogram may be seen. As it could be expected, the changes in the Spectrogram are related with changes in the PARCORgram. In many applications, such as Phone Labelling [Robinson.94], or Formant Tracing [Nagayama.94], it should be desirable to detect the set of Vector Parameters responsible for these changes, or to inferr Formant Dynamics from Vector Parameters. Through the present work one such application using Gradient-Adaptive Lattices and Time-Delay Neural Networks will be described.

Estimation of speech formant-dynamics using neural networks.

We present here a new technique for making predictions on recommender systems based on collaborative filtering. The underlying idea is based on selecting a different number of neighbors for each user, instead of, as it is usually made, selecting always a constant number k of neighbors. In this way, we have improved significantly the accuracy of the recommender systems.

Collaborative Filtering Based on Choosing a Different Number of Neighbors for Each User

Recommender systems are widely used by companies that sell all or some of their products via the Internet. Furthermore, they are destined to take on an even more important role when their use is generalized as a Web 2.0 social service and is no longer only linked to e-commerce companies. The recommendations that a recommender system offers any given user are based on the preferences shown by a given group of users that have been selected with his/her own similarities. In this paper, we present a series of equations that enable us to obtain each user's importance according to the quality of the recommendations he/she receives and the quality of the recommendations he/she generates. In order to demonstrate the correct operation of the proposed method, we have designed and carried out 90 comparative experiments based on the MovieLens database, whereby we have obtained results that improve the performance of the recommender system at the same time as they increase its levels of accuracy. Each user's values of importance can be used for the following: to restrict or increase the number of recommendations provided to a user, to add information about the reliability of the suggested recommendations, to inform about the level of influence a user has at each time on the recommendations he/she contributes to others, to achieve an objective measurement in order to reward or encourage users with higher levels of importance and even to make it possible to design and implement applications that enable the recommendations made to be monitored and optimized.

Recommender systems: Improving collaborative filtering results

Recommender system (RS) clustering is an important issue, both for the improvement of the collaborative filtering (CF) accuracy and to obtain analytical information from their high sparse datasets. RS items and users usually share features belonging to different clusters, e.g., a musical-comedy movie. Soft clustering, therefore, is the CF clustering’s most natural approach. In this paper, we propose a new prediction approach for probabilistic soft clustering methods. In addition, we put to test a not traditional scientific documentation CF dataset: SD4AI, and we compare results with the MovieLens baseline. Not traditional CF datasets have challenging features, such as not regular rating frequency distributions, broad range of rating values, and a particularly high sparsity. The results show the suitability of using soft-clustering approaches, where their probabilistic overlapping parameters find optimum values when balanced hard/soft clustering is used. This paper opens some promising lines of research, such as RSs’ use in the scientific documentation field, the Internet of Things-based datasets processing, and design of new model-based soft clustering methods.

/pdf/a-new-recommendation-approach-based-on-probabilistic-soft-cyyedb0y7z.pdf

A New Recommendation Approach Based on Probabilistic Soft Clustering Methods: A Scientific Documentation Case Study

Spectra is usually shown as a two-dimensional graph where colors are directly related to signal levels. A great deal of speech recognition work and research takes this type of parameter directly. In this paper we propose to combine typical signal level values with the vectorial components of a Slope matrix containing orientation information on spectra surfaces. This additional information will enable us to obtain an enhanced speech signal spectra as well as formant evolution detection and a matching method to compare speech spectra sections. The mathematical formalization is based on vector analysis and matrix operations, where the basic components are the normal vectors to a set of triangular surfaces covering the spectral values. This formalism enables the use of mathematical tools (Matlab or similar) in a very easy way; and from here it is possible to program algorithms and visualize the results efficiently.

Jesús Bobadilla

Papers

Estimation of speech formant-dynamics using neural networks.

Collaborative Filtering Based on Choosing a Different Number of Neighbors for Each User

Recommender systems: Improving collaborative filtering results

A New Recommendation Approach Based on Probabilistic Soft Clustering Methods: A Scientific Documentation Case Study

New Speech Enhancement Approach for Formant Evolution Detection