526 publications from this institution
For more than a century, Markov chain models have had a tremendous effect on research and industry alike in various domains such as physics, chemistry, biology, and computer science. However, limited research has been performed on how to incorporate contextual conditions into the modeling phase. Existing approaches are performed in an ad hoc manner or fail to consider multiple contextual dimensions simultaneously. Inspired by the main paradigms in context-aware recommender systems, we suggest five novel approaches for learning contextual Markov chain models. In particular we suggest three contextual pre-filtering models that operate by learning multiple sub-models from various data partitions, each tailored to specific contextual aspects; a contextual model that directly extents the Markovian model with the contextual features; and lastly, a post-filtering model, which serves to refine the predictions made by a traditional non-contextual model by considering the available contextual information. We evaluate the suggested methods in two use cases: analysis of web browsing activities, and attack propagation patterns in honeypot systems. Experiments with massive datasets that contain millions of records indicate that taking the context of modeled sequences into consideration can substantially improve the accuracy of predicting and ranking the next possible element in such sequences. Moreover, we demonstrate the high scalability of the suggested methods, making them suitable for analyzing a vast amount of sequential data.
Decision trees have become one of the most powerful and popular approaches in knowledge discovery and data mining; it is the science of exploring large and complex bodies of data in order to discover useful patterns. Decision tree learning continues to evolve over time. Existing methods are constantly being improved and new methods introduced. This 2nd Edition is dedicated entirely to the field of decision trees in data mining; to cover all aspects of this important technique, as well as improved or new methods and techniques developed after the publication of our first edition. In this new edition, all chapters have been revised and new topics brought in. New topics include Cost-Sensitive Active Learning, Learning with Uncertain and Imbalanced Data, Using Decision Trees beyond Classification Tasks, Privacy Preserving Decision Tree Learning, Lessons Learned from Comparative Studies, and Learning Decision Trees for Big Data. A walk-through guide to existing open-source data mining software is also included in this edition. This book invites readers to explore the many benefits in data mining that decision trees offer: Self-explanatory and easy to follow when compacted Able to handle a variety of input data: nominal, numeric and textual Scales well to big data Able to process datasets that may have errors or missing values High predictive performance for a relatively small computational effort Available in many open source data mining packages over a variety of platforms Useful for various tasks, such as classification, regression, clustering and feature selection Readership: Researchers, graduate and undergraduate students in information systems, engineering, computer science, statistics and management.
Online Social Networks (OSNs), such as Facebook and Twitter, have become an integral part of our daily lives. There are hundreds of OSNs, each with its own focus in that each offers particular services and functionalities. Recent studies show that many OSN users create several accounts on multiple OSNs using the same or different personal information. Collecting all the available data of an individual from several OSNs and fusing it into a single profile can be useful for many purposes. In this paper, we introduce novel machine learning based methods for solving Entity Resolution (ER), a problem for matching user profiles across multiple OSNs. The presented methods are able to match between two user profiles from two different OSNs based on supervised learning techniques, which use features extracted from each one of the user profiles. By using the extracted features and supervised learning techniques, we developed classifiers which can perform entity matching between two profiles for the following scenarios: (a) matching entities across two OSNs; (b) searching for a user by similar name; and (c) de-anonymizing a user's identity. The constructed classifiers were tested by using data collected from two popular OSNs, Facebook and Xing. We then evaluated the classifiers' performances using various evaluation measures, such as true and false positive rates, accuracy, and the Area Under the receiver operator Curve (AUC). The constructed classifiers were evaluated and their classification performance measured by AUC was quite remarkable, with an AUC of up to 0.982 and an accuracy of up to 95.9% in identifying user profiles across two OSNs.
The MediaScout system is envisioned to function as personalized media (audio, video, print) service within mobile phones, online media portals, sling boxes, etc. The MediaScout recommender engine uses a novel stereotype-based recommendation engine. Upon the registration of new users the system must decide how to classify the new users to existing stereotypes. In this paper we present a method to achieve this classification through an anytime, interactive questionnaire, created automatically upon the generation of new stereotypes. A comparative study performed on the IMDB database illustrates the advantages of the new system
Context-aware systems enable the sensing and analysis of user context in order to provide personalised services. Our study is part of growing research efforts examining how high-dimensional data collected from mobile devices can be utilised to infer users' dynamic preferences that are learned over time. We suggest novel methods for inferring the category of the item liked in a specific contextual situation, by applying encoder-decoder learners (long short-term memory networks and auto encoders) on mobile sensor data. In these approaches, the encoder-decoder learners reduce the dimensionality of the contextual features to a latent representation which is learned over time. Given new contextual sensor data from a user, the latent patterns discovered from each deep learner is used to predict the liked item's category in the given context. This can greatly enhance a variety of services, such as mobile online advertising and context-aware recommender systems. We demonstrate our contribution with a point of interest (POI) recommender system in which we label contextual situations with the items' categories. Empirical results utilising a real world data set of contextual situations derived from mobile phones sensors log show a significant improvement (up to 73% improvement) in prediction accuracy compared with state of the art classification methods.
We propose a stealthy and powerful backdoor attack on neural networks based on data poisoning. In contrast to previous attacks, both the poison and the trigger in our method are stealthy. We are able to change the model’s classification of samples from a source class to a target class chosen by the attacker. We do so by using a small number of poisoned training samples with nearly imperceptible perturbations, without changing their labels. At inference time, we use a stealthy perturbation added to the attacked samples as a trigger. This perturbation is crafted as a universal adversarial perturbation (UAP), and the poison is crafted using gradient alignment coupled to this trigger. Our method is highly efficient in crafting time compared to previous methods and requires only a trained surrogate model without additional retraining. Our attack achieves state-of-the-art results in terms of attack success rate while maintaining high accuracy on clean samples.
Recommendation systems often compute fixed‐length lists of recommended items to users. Forcing the system to predict a fixed‐length list for each user may result in different confidence levels for the computed recommendations. Reporting the system's confidence in its predictions (the recommendation strength) can provide valuable information to users in making their decisions. In this article, we investigate several different displays of a system's confidence to users and conclude that some displays are easier to understand and are favored by most users. We continue to investigate the effect confidence has on users in terms of their perception of the recommendation quality and the user experience with the system. Our studies show that it is not easier for users to identify relevant items when confidence is displayed. Still, users appreciate the displays and trust them when the relevance of items is difficult to establish.
No abstract is provided for this article.
No abstract is provided for this article.
No abstract is provided for this article.
The query-performance prediction task aims at estimating the retrieval effectiveness of queries without obtaining relevance feedback from users. Most of the recently proposed predictors were empirically evaluated with various datasets to demonstrate their merits. We propose a framework for theoretical categorization and estimation of the value of query performance predictors (QPP) without empirical evaluation. We demonstrate the application of the proposed framework on four representative selected predictors and show how it emphasizes their strengths and weaknesses. The main contribution of this work is the theoretical grounded categorization of representative QPP.
Driving under the influence of alcohol is a widespread phenomenon in the US where it is considered a major cause of fatal accidents. In this research we present a novel approach and concept for detecting intoxication from motion differences obtained by the sensors of wearable devices. We formalize the problem of drunkenness detection as a supervised machine learning task, both as a binary classification problem (drunk or sober) and a regression problem (the breath alcohol content level). In order to test our approach, we collected data from 30 different subjects (patrons at three bars) using Google Glass and the LG G-watch, Microsoft Band, and Samsung Galaxy S4. We validated our results against an admissible breathalyzer used by the police. A system based on this concept, successfully detected intoxication and achieved the following results: 0.95 AUC and 0.05 FPR, given a fixed TPR of 1.0. Applications based on our system can be used to analyze the free gait of drinkers when they walk from the car to the bar and vice-versa, in order to alert people, or even a connected car and prevent people from driving under the influence of alcohol.
Online health communities are a major source for patients and their informal caregivers in the process of gathering information and seeking social support. The Cancer Survivors Network of the American Cancer Society has many users and presents a large number of user interactions with regards to coping with cancer. Sentiment analysis is an important process in understanding members' needs and concerns and the impact of users' responses on other members. It aims to determine the participants' subjective attitude and reflect their emotions. Analyzing the sentiment of posts in online health communities enables the investigation of various factors such as what affects the sentiment change and discovery of sentiment change patterns. Since each writer has his or her own personality, and temporal emotional state, behavioral traits can be reflected in the writer's writing style. Pronouns are function-words which often convey some unique styling patterns into the texts. Drawing on a lexical approach to emotions, we conduct factor analysis on the use of pronouns in self-descriptions texts. Our analysis shows that the usage of pronouns has an effect on sentiment classification. Moreover, we evaluated the use of pronouns in our domain, and found it different than standard English usage.
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 by achieving better coverage of labels and inter-label correlations. 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 to the RAKEL algorithm and to other state-of-the-art algorithms.
