Abstract
Machine learning processes consist in collecting data, obtaining a model and applying it to a given task. Given a new task, the standard approach is to restart the learning process and obtain a new model. However, previous learning experience can be exploited to assist the new learning process. The two most studied approaches for this are meta-learning and transfer learning. Metalearning can be used for selecting the predictive model to use on a new dataset. Transfer learning allows the reuse of knowledge from previous tasks. However, when multiple heterogeneous tasks are available as potential sources for transfer, the question is which one to use. One approach to address this problem is metalearning. In this paper we investigate the feasibility of this approach. We propose a method to transfer weights from a source trained neural network to initialize a network that models a potentially very different target dataset. Our experiments with 14 datasets indicate that this method enables faster convergence without significant difference in accuracy provided that the source task is adequately chosen. This means that there is potential for applying metalearning to support transfer between heterogeneous datasets.

Abstract
The Oil and Gas Exploration & Production (E&P) field deals with high-dimensional heterogeneous data, collected at different stages of the E&P activities from various sources. Over the years different soft-computing algorithms have been proposed for data-driven oil and gas applications. The most popular by far are Artificial Neural Networks, but there are applications of Fuzzy Logic systems, Support Vector Machines, and Evolutionary Algorithms (EAs) as well. This article provides an overview of the applications of EAs in the oil and gas E&P industry. The relevant literature is reviewed and categorised, showing an increasing interest amongst the geoscience community. © 2016 ACM.

Abstract
With the emergence of online Music Streaming Services (MSS) such as Pandora and Spotify, listening to music online became very popular. Despite the availability of these services, users face the problem of finding among millions of music tracks the ones that match their music taste. MSS platforms generate interaction data such as users' defined playlists enriched with relevant metadata. These metadata can be used to predict users' preferences and facilitate personalized music recommendation. In this work, we aim to infer music tastes of users by using personal playlist information. Characterizing users' taste is important to generate trustable recommendations when the amount of usage data is limited. Here, we propose to predict the users' preferred music feature's value (e.g. Genre as a feature has different values like P op, Rock, etc.) by modeling, not only usage information, but also music description features. Music attribute information and usage data are typically dealt with separately. Our method FPMF (Feature Prediction based on Matrix Factorization) treats music feature values as virtual users and retrieves the preferred feature values for real target users. Experimental results indicate that our proposal is able to handle the item cold start problem and can retrieve preferred music feature values with limited usage data. Furthermore, our proposal can be useful in recommendation explanation scenarios. © 2016 ACM.

Abstract
In the web environment, most of the queries issued by users are implicit by nature. Inferring the different temporal intents of this type of query enhances the overall temporal part of the web search results. Previous works tackling this problem usually focused on news queries, where the retrieval of the most recent results related to the query are usually sufficient to meet the user's information needs. However, few works have studied the importance of time in queries such as "Philip Seymour Hoffman" where the results may require no recency at all. In this work, we focus on this type of queries named "time-sensitive queries" where the results are preferably from a diversified time span, not necessarily the most recent one. Unlike related work, we follow a content-based approach to identify the most important time periods of the query and integrate time into a re-ranking model to boost the retrieval of documents whose contents match the query time period. For that purpose, we define a linear combination of topical and temporal scores, which reflects the relevance of any web document both in the topical and temporal dimensions, thus contributing to improve the effectiveness of the ranked results across different types of queries. Our approach relies on a novel temporal similarity measure that is capable of determining the most important dates for a query, while filtering out the non-relevant ones. Through extensive experimental evaluation over web corpora, we show that our model offers promising results compared to baseline approaches. As a result of our investigation, we publicly provide a set of web services and a web search interface so that the system can be graphically explored by the research community.

Abstract
The aim of this work is to develop an application for Android able to classifying urban sounds in a real life context. It also enables the collection and classification of new sounds. To train our classifier we use the UrbanSound8K data set available online. We have used a hybrid approach to obtain features, by combining SAX-based multiresolution motif discovery with Mel-Frequency Cepstral Coefficients (MFCC). We also describe different configurations of motif discovery for defining attributes and compare the use of Random Forest and SVM algorithms on this kind of data. Copyright 2016 ACM.

Abstract
Online recommender systems often deal with continuous, potentially fast and unbounded ows of data. Ensemble methods for recommender systems have been used in the past in batch algorithms, however they have never been studied with incremental algorithms, that are capable of processing those data streams on the y. We propose online bagging, using an incremental matrix factorization algorithm for positiveonly data streams. Using prequential evaluation, we show that bagging is able to improve accuracy more than 20% over the baseline with small computational overhead.