526 publications from this institution
Traditionally user authentication is based on a username and password. However, a logged station is still vulnerable to imposters when the user leaves her computer without logging-off. Keystroke dynamics methods can be useful for continuously verifying a user once the authentication process has successfully ended. However, current methods require long sessions and significant amounts of keystrokes to reliably verify users. We propose a new method that compactly represents the keystroke patterns by joining similar pairs of consecutive keystrokes. This automatically created representation reduces the session size required for inducing the user's verification model. The proposed method was evaluated on 21 legitimate users and 165 attackers. The results were encouraging and suggest that the detection performance of the proposed method is better than that of existing methods. Specifically we attained a false acceptance rate (FAR) of 3.47% and false rejection rate (FRR) of 0% using only 250 keystrokes.
Researchers from various disciplines such as pattern recognition, statistics, and machine learning have explored the use of ensemble methodology since the late seventies. Thus, they are faced with a wide variety of methods, given the growing interest in the field. This book aims to impose a degree of order upon this diversity by presenting a coherent and unified repository of ensemble methods, theories, trends, challenges and applications. The book describes in detail the classical methods, as well as the extensions and novel approaches developed recently. Along with algorithmic descriptions of each method, it also explains the circumstances in which this method is applicable and the consequences and the trade-offs incurred by using the method.
No abstract is provided for this article.
In this paper we present a new approach to content-based transfer learning for solving the data sparsity problem in cases when the users' preferences in the target domain are either scarce or unavailable, but the necessary information on the preferences exists in another domain. We show that training a system to use such information across domains can produce better performance. Specifically, we represent users' behavior patterns based on topological graph structures. Each behavior pattern represents the behavior of a set of users, when the users' behavior is defined as the items they rated and the items' rating values. In the next step we find a correlation between behavior patterns in the source domain and behavior patterns in the target domain. This mapping is considered a bridge between the two domains. Based on the correlation and content-attributes of the items, we train a machine learning model to predict users' ratings in the target domain. When we compare our approach to the popularity approach and KNN-cross-domain on a real world dataset, the results show that on an average of 83$%$ of the cases our approach outperforms both methods.
Recent developments in large language models (LLMs) trained on large-scale unstructured textual data have produced high-achieving models. However, it remains a challenge to train an LLM on vast structured (tabular) data for the task of understanding the information captured in the data and answering questions regarding the data. We propose a novel method - TQwerE, for approximating SQL aggregated queries’ results over large data sets. Our main focus was to reduce query latency and incurred costs. Moreover, since we focus on large data sets, majority of models that scan raw data are not applicable. Instead, our method fine tunes Jurassic-2 to learn the relations between aggregated SQL queries and their results without referring directly to the underlying raw data. We demonstrate TQwerE’s ability to approximate aggregated queries with state-of-the-art accuracy and speed. We evaluated TQwerE on twelve datasets, and our results demonstrated its superiority to both the state-of-the-art methods.
No abstract is provided for this article.
In this paper we examine the effect of applying ensemble learning to the performance of collaborative filtering methods. We present several systematic approaches for generating an ensemble of collaborative filtering models based on a single collaborative filtering algorithm (single-model or homogeneous ensemble). We present an adaptation of several popular ensemble techniques in machine learning for the collaborative filtering domain, including bagging, boosting, fusion and randomness injection. We evaluate the proposed approach on several types of collaborative filtering base models: k- NN, matrix factorization and a neighborhood matrix factorization model. Empirical evaluation shows a prediction improvement compared to all base CF algorithms. In particular, we show that the performance of an ensemble of simple (weak) CF models such as k-NN is competitive compared with a single strong CF model (such as matrix factorization) while requiring an order of magnitude less computational cost.
No abstract is provided for this article.
Ensemble methods have been shown to be an effective tool for solving multi-label classification tasks. In the RAndom k-labELsets (RAKEL) algorithm, each member of the ensemble is associated with a small randomly-selected subset of k labels. Then, a single label classifier is trained according to each combination of elements in the subset. In this paper we adopt a similar approach, however, instead of randomly choosing subsets, we select the minimum required subsets of k labels that cover all labels and meet additional constraints such as coverage of inter-label correlations. Construction of the cover is achieved by formulating the subset selection as a minimum set covering problem (SCP) and solving it by using approximation algorithms. Every cover needs only to be prepared once by offline algorithms. Once prepared, a cover may be applied to the classification of any given multi-label dataset whose properties conform with those of the cover. The contribution of this paper is two-fold. First, we introduce SCP as a general framework for constructing label covers while allowing the user to incorporate cover construction constraints. We demonstrate the effectiveness of this framework by proposing two construction constraints whose enforcement produces covers that improve the prediction performance of random selection. Second, we provide theoretical bounds that quantify the probabilities of random selection to produce covers that meet the proposed construction criteria. The experimental results indicate that the proposed methods improve multi-label classification accuracy and stability compared with the RAKEL algorithm and to other state-of-the-art algorithms.
No abstract is provided for this article.
Decision forests are considered the best practice in many machine learning challenges, mainly due to their superior predictive performance. However, simple models like decision trees may be preferred over decision forests in cases in which the generated predictions must be efficient or interpretable (e.g. in insurance or health-related use cases). This paper presents a novel method for transforming a decision forest into an interpretable decision tree, which aims at preserving the predictive performance of decision forests while enabling efficient classifications that can be understood by humans. This is done by creating a set of rule conjunctions that represent the original decision forest; the conjunctions are then hierarchically organized to form a new decision tree. We evaluate the proposed method on 33 UCI datasets and show that the resulting model usually approximates the ROC AUC gained by random forest while providing an interpretable decision path for each classification.
Decision forest is an ensemble classification method that combines multiple decision trees to in a manner that results in more accurate classifications. By combining multiple heterogeneous decision trees, decision forest is effective in mitigating noise that is often prevalent in real-world classification tasks. This paper presents a new genetic algorithm for constructing a decision forest. Each decision tree classifier is trained using a disjoint set of attributes. Moreover, we examine the effectiveness of using a Vapnik–Chervonenkis dimension bound for evaluating the fitness function of decision forest. The new algorithm was tested on various datasets. The obtained results have been compared to other methods, indicating the superiority of the proposed algorithm. © 2008 Wiley Periodicals, Inc.
