Learning to Rank (L2R) is the core task of many Information Retrieval systems. Recently, a great effort has been put on exploring Deep Neural Networks (DNNs) for L2R, with significant results. However, risk-sensitiveness, an important and recent advance in the L2R arena, that reduces variability and increases trust, has not been incorporated into Deep Neural L2R yet. Risk-sensitive measures are important to assess the risk of an IR system to perform worse than a set of baseline IR systems for several queries. However, the risk-sensitive measures described in the literature have a non-smooth behavior, making them difficult, if not impossible, to be optimized by DNNs. In this work we solve this difficult problem by proposing a family of new loss functions -- \riskloss\ -- that support a smooth risk-sensitive optimization. \riskloss\ introduces two important contributions: (i) the substitution of the traditional NDCG or MAP metrics in risk-sensitive measures with smooth loss functions that evaluate the correlation between the predicted and the true relevance order of documents for a given query and (ii) the use of distinct versions of the same DNN architecture as baselines by means of a multi-dropout technique during the smooth risk-sensitive optimization, avoiding the inconvenience of assessing multiple IR systems as part of DNN training. We empirically demonstrate significant achievements of the proposed \riskloss\ functions when used with recent DNN methods in the context of well-known web-search datasets such as WEB10K, YAHOO, and MQ2007. Our solutions reach improvements of 8% in effectiveness (NDCG) while improving in around 5% the risk-sensitiveness (\grisk\ measure) when applied together with a state-of-the-art Self-Attention DNN-L2R architecture. Furthermore, \riskloss\ is capable of reducing by 28% the losses over the best evaluated baselines and significantly improving over the risk-sensitive state-of-the-art non-DNN method (by up to 13.3%) while keeping (or even increasing) overall effectiveness. All these results ultimately establish a new level for the state-of-the-art on risk-sensitiveness and DNN-L2R research.

Bee-mediated pollination greatly increases the size and weight of tomato fruits. Therefore, distinguishing between the local set of bees-those that are efficient pollinators-is essential to improve the economic returns for farmers. To achieve this, it is important to know the identity of the visiting bees. Nevertheless, the traditional taxonomic identification of bees is not an easy task, requiring the participation of experts and the use of specialized equipment. Due to these limitations, the development and implementation of new technologies for the automatic recognition of bees become relevant. Hence, we aim to verify the capacity of Machine Learning (ML) algorithms in recognizing the taxonomic identity of visiting bees to tomato flowers based on the characteristics of their buzzing sounds. We compared the performance of the ML algorithms combined with the Mel Frequency Cepstral Coefficients (MFCC) and with classifications based solely on the fundamental frequency, leading to a direct comparison between the two approaches. In fact, some classifiers powered by the MFCC-especially the SVM-achieved better performance compared to the randomized and sound frequency-based trials. Moreover, the buzzing sounds produced during sonication were more relevant for the taxonomic recognition of bee species than analysis based on flight sounds alone. On the other hand, the ML classifiers performed better in recognizing bees genera based on flight sounds. Despite that, the maximum accuracy obtained here (73.39% by SVM) is still low compared to ML standards. Further studies analyzing larger recording samples, and applying unsupervised learning systems may yield better classification performance. Therefore, ML techniques could be used to automate the taxonomic recognition of flower-visiting bees of the cultivated tomato and other buzz-pollinated crops. This would be an interesting option for farmers and other professionals who have no experience in bee taxonomy but are interested in improving crop yields by increasing pollination.

Recent advances in text-related tasks on the Web, such as text (topic) classification and sentiment analysis, have been made possible by exploiting mostly the "rule of more": more data (massive amounts) more computing power, more complex solutions. We propose a shift in the paradigm to do "more with less" by focusing, at maximum extent, just on the task at hand (e.g., classify a single test instance). Accordingly, we propose MetaLazy, a new supervised lazy text classification meta-strategy that greatly extends the scope of lazy solutions. Lazy classifiers postpone the creation of a classification model until a given test instance for decision making is given. MetaLazy exploits new ideas and solutions, which have in common their lazy nature, producing altogether a solution for text classification, which is simpler, more efficient, and less data demanding than new alternatives. It extends and evolves the lazy creation of the model for the test instance by allowing: (i) to dynamically choose the best classifier for the task; (ii) the exploration of distances in the neighborhood of the test document when learning a classification model, thus diminishing the importance of irrelevant training instances; and (iii) a better representational space for training and test documents by augmenting them, in a lazy fashion, with new co-occurrence based features considering just those observed in the specific test instance. In a sizeable experimental evaluation, considering topics and sentiment analysis datasets and nine baselines, we show that our MetaLazy instantiations are among the top performers in most situations, even when compared to state-of-the-art deep learning classifiers such as Deep Network Transformer Architectures.

We propose new solutions that enhance and extend the already very successful application of meta-features to text classification. Our newly proposed meta-features are capable of: (1) improving the correlation of small pieces of evidence shared by neighbors with labeled categories by means of synthetic document representations and (local and global) hyperplane distances; and (2) estimating the level of error introduced by these newly proposed and the existing meta-features in the literature, specially for hard-to-classify regions of the feature space. Our experiments with large and representative number of datasets show that our new solutions produce the best results in all tested scenarios, achieving gains of up to 12% over the strongest meta-feature proposal of the literature.

In this paper, we advance the state-of-the-art in topic modeling by means of a new document representation based on pre-trained word embeddings for non-probabilistic matrix factorization. Specifically, our strategy, called CluWords, exploits the nearest words of a given pre-trained word embedding to generate meta-words capable of enhancing the document representation, in terms of both, syntactic and semantic information. The novel contributions of our solution include: (i)the introduction of a novel data representation for topic modeling based on syntactic and semantic relationships derived from distances calculated within a pre-trained word embedding space and (ii)the proposal of a new TF-IDF-based strategy, particularly developed to weight the CluWords. In our extensive experimentation evaluation, covering 12 datasets and 8 state-of-the-art baselines, we exceed (with a few ties) in almost cases, with gains of more than 50% against the best baselines (achieving up to 80% against some runner-ups). Finally, we show that our method is able to improve document representation for the task of automatic text classification.

Summary Learning to rank (L2R) works by constructing a ranking model from training data so that, given a new query, the model is able to generate an effective rank of the objects for the query. Almost all work in L2R focus on ranking accuracy leaving performance and scalability overlooked. However, performance is a critical factor, especially when dealing with on‐demand queries. In this scenario, Learning to Rank using association rules has been shown to be extremely effective but only at a high computational cost. In this work, we show how to exploit parallelism on rule‐based systems to: i) drastically reduce L2R training datasets using selective sampling and ii) to generate query customized ranking models on the fly. We present parallel algorithms and GPU implementations for these two tasks showing that dataset reduction takes only a few seconds with speedups up to 148x over a serial baseline, and that queries can be processed in only a few milliseconds with speedups of 1000x over a serial baseline and 29x over a parallel baseline for the best case. We also extend the implementations to work with multiple GPUs, further increasing the speedup over the baselines and showing the scalability of our proposed algorithms.

Learning to Rank (L2R) is one of the main research lines in Information Retrieval. Risk-sensitive L2R is a sub-area of L2R that tries to learn models that are good on average while at the same time reducing the risk of performing poorly in a few but important queries (e.g., medical or legal queries). One way of reducing risk in learned models is by selecting and removing noisy, redundant features, or features that promote some queries to the detriment of others. This is exacerbated by learning methods that usually maximize an average metric (e.g., mean average precision (MAP) or Normalized Discounted Cumulative Gain (NDCG)). However, historically, feature selection (FS) methods have focused only on effectiveness and feature reduction as the main objectives. Accordingly, in this work, we propose to evaluate FS for L2R with an additional objective in mind, namely risk-sensitiveness . We present novel single and multi-objective criteria to optimize feature reduction, effectiveness, and risk-sensitiveness, all at the same time. We also introduce a new methodology to explore the search space, suggesting effective and efficient extensions of a well-known Evolutionary Algorithm (SPEA2) for FS applied to L2R. Our experiments show that explicitly including risk as an objective criterion is crucial to achieving a more effective and risk-sensitive performance. We also provide a thorough analysis of our methodology and experimental results.

The analysis of affinity or similarity between people is an important task in the study of social dynamics. Traditional methods for determining similarity depends on considerable amount of data regarding people's preferences and features. Those methods present limitations when the data is scarce and/or changes constantly. This paper introduces a new method for determining people similarity that does not suffer from the same problems. The method can learn a customized similarity function based on social variables of friends that attend the same events (concerts, parties, conferences etc), collected from social networks. Two types of optimization algorithms for learning a similarity function are presented: The universal function approximator modelling, which relays on the relationship of social attributes and a friends' importance ranking; and the populational evolutionary modelling, which linearly combines social variables. Both models were tested in a generalist and in a specialist approach. The results show that the specialist approach exceeded in almost 38 % the generalist approach using populational evolutionary methods and in almost 69 % when using the universal function approximator methods. Among the implemented optimization algorithms employed inside the methods for learning similarity, Genetic Algorithm and Particle Swarm Optimization presented better performance for the populational evolutionary methods and the Artificial Neural Network presented the best performance overall using the universal function approximator modelling.

We address the problem of automatically learning to classify texts by exploiting information derived from meta-features, i.e., features derived from the original bag-of-words representation. Specifically, we provide an in-depth analysis on the recently proposed distance-based meta-features, a data engineering technique that relies on the distance between documents to transform the original feature space into a new one, potentially smaller and more informed. Despite its potential, the meta-feature space may be unnecessarily complex and highly dimensional, which increases the tendency of overfitting, limits the application of meta-features in different contexts, and increases computational costs. In this work, we propose the use of multi-objective strategies to reduce the number of meta-features while maximizing the classification effectiveness, when considering the adequacy of the selected meta-features to a particular dataset or classification method. We present effective and efficient proposals for meta-feature selection that can substantially reduce the number of meta-features by up to 89 percent while keeping or improving the classification effectiveness, something not possible with any of the evaluated baselines. We also use our selection strategies as evaluation tools to analyze different combinations of meta-features. We found very compact combinations of meta-features that can achieve high classification effectiveness in most datasets, despite their peculiarities.

Learning to Rank (L2R) is currently an essential task in basically all types of information systems given the huge and ever increasing amount of data made available. While many solutions have been proposed to improve L2R functions, relatively little attention has been paid to the task of improving the quality of the feature space. L2R strategies usually rely on dense feature representations, which contain noisy or redundant features, increasing the cost of the learning process, without any benefits. Although feature selection (FS) strategies can be applied to reduce dimensionality and noise, side effects of such procedures have been neglected, such as the risk of getting very poor predictions in a few (but important) queries. In this paper we propose multi-objective FS strategies that optimize both aspects at the same time: ranking performance and risk-sensitive evaluation. For this, we approximate the Pareto-optimal set for multi-objective optimization in a new and original application to L2R. Our contributions include novel FS methods for L2R which optimize multiple, potentially conflicting, criteria. In particular, one of the objectives (risk-sensitive evaluation) has never been optimized in the context of FS for L2R before. Our experimental evaluation shows that our proposed methods select features that are more effective (ranking performance) and low-risk than those selected by other state-of-the-art FS methods.

The unprecedented growth of available data nowadays has stimulated the development of new methods for organizing and extracting useful knowledge from this immense amount of data. Automatic Document Classification (ADC) is one of such methods, that uses machine learning techniques to build models capable of automatically associating documents to well-defined semantic classes. ADC is the basis of many important applications such as language identification, sentiment analysis, recommender systems, spam filtering, among others. Recently, the use of meta-features has been shown to substantially improve the effectiveness of ADC algorithms. In particular, the use of meta-features that make a combined use of local information (through kNN-based features) and global information (through category centroids) has produced promising results. However, the generation of these meta-features is very costly in terms of both, memory consumption and runtime since there is the need to constantly call the kNN algorithm. We take advantage of the current manycore GPU architecture and present a massively parallel version of the kNN algorithm for highly dimensional and sparse datasets (which is the case for ADC). Our experimental results show that we can obtain speedup gains of up to 15x while reducing memory consumption in more than 5000x when compared to a state-of-the-art parallel baseline. This opens up the possibility of applying meta-features based classification in large collections of documents, that would otherwise take too much time or require the use of an expensive computational platform.

With the development of web 2.0, social networks have achieved great space on the internet, with that many users provide information and interests about themselves. There are expert systems that use the user's interests to recommend different products, these systems are known as Recommender Systems. One of the main techniques of a Recommender Systems is the Collaborative Filtering (User based) which recommends products to users based on what other similar people liked in the past. However, the methods to determine similarity between users have presented some problems. Therefore, this work presents a proposal of using social variables in the composition of the similarity function applied to a user on the recommendation of events. To test the proposal, details of friends and events of two target-users of the social network Facebook have been extracted. The results were compared with different deterministic heuristics, the Euclidean Distance and a aleatory method. The proposed model showed promising results and great potential to expand to different contexts.

Recommendation systems work as a counselor, behaving in such a way to guide people in the discovery of products of interest. There are various techniques and approaches in the literature that enable generating recommendations. This is interesting because it emphasizes the diversity of options; on the other hand, it can cause doubt to the system designer about which is the best technique to use. Each of these approaches has particularities and depends on the context to be applied. Thus, the decision to choose among techniques become complex to be done manually. This article proposes an evolutionary approach for combining results of recommendation techniques in order to automate the choice of techniques and get fewer errors in recommendations. To evaluate the proposal, experiments were performed with a dataset from MovieLens and some of Collaborative Filtering techniques. The results show that the combining methodology proposed in this paper performs better than any one of collaborative filtering technique separately in the context addressed. The improvement varies from 9.02% to 48.21% depending on the technique and the experiment executed.

This paper addresses the problem of automatically learning to classify texts by exploiting information derived from meta-level features (i.e., features derived from the original bag-of-words representation). We propose new meta-level features derived from the class distribution, the entropy and the within-class cohesion observed in the k nearest neighbors of a given test document x, as well as from the distribution of distances of x to these neighbors. The set of proposed features is capable of transforming the original feature space into a new one, potentially smaller and more informed. Experiments performed with several standard datasets demonstrate that the effectiveness of the proposed meta-level features is not only much superior than the traditional bag-of-word representation but also superior to other state-of-art meta-level features previously proposed in the literature. Moreover, the proposed meta-features can be computed about three times faster than the existing meta-level ones, making our proposal much more scalable. We also demonstrate that the combination of our meta features and the original set of features produce significant improvements when compared to each feature set used in isolation.

BACKGROUND: Phylogenetic comparative analyses usually rely on a single consensus phylogenetic tree in order to study evolutionary processes. However, most phylogenetic trees are incomplete with regard to species sampling, which may critically compromise analyses. Some approaches have been proposed to integrate non-molecular phylogenetic information into incomplete molecular phylogenies. An expanded tree approach consists of adding missing species to random locations within their clade. The information contained in the topology of the resulting expanded trees can be captured by the pairwise phylogenetic distance between species and stored in a matrix for further statistical analysis. Thus, the random expansion and processing of multiple phylogenetic trees can be used to estimate the phylogenetic uncertainty through a simulation procedure. Because of the computational burden required, unless this procedure is efficiently implemented, the analyses are of limited applicability. RESULTS: In this paper, we present efficient algorithms and implementations for randomly expanding and processing phylogenetic trees so that simulations involved in comparative phylogenetic analysis with uncertainty can be conducted in a reasonable time. We propose algorithms for both randomly expanding trees and calculating distance matrices. We made available the source code, which was written in the C++ language. The code may be used as a standalone program or as a shared object in the R system. The software can also be used as a web service through the link: http://purl.oclc.org/NET/sunplin/. CONCLUSION: We compare our implementations to similar solutions and show that significant performance gains can be obtained. Our results open up the possibility of accounting for phylogenetic uncertainty in evolutionary and ecological analyses of large datasets.

Due to the increasing amount of information present on the Web, Automatic Document Classification (ADC) has become an important research topic. ADC usually follows a standard supervised learning strategy, where we first build a model using preclassified documents and then use it to classify new unseen documents. One major challenge for ADC in many scenarios is that the characteristics of the documents and the classes to which they belong may change over time. However, most of the current techniques for ADC are applied without taking into account the temporal evolution of the collection of documents