Olfa Nasraoui

Verified

Olfa verified their affiliation via an institutional email.

Learn more

Verified

Olfa verified their affiliation via an institutional email.

Learn more

• PhD. in Computer Engineering and Computer Science
• Professor (Full) at University of Louisville

In this paper, we present a complete framework and findings in mining Web usage patterns from Web log files of a real Web site that has all the challenging aspects of real-life Web usage mining, including evolving user profiles and external data describing an ontology of the Web content. Even though the Web site under study is part of a nonprofit o...

Most accurate recommender systems are black-box models, hiding the reasoning behind their recommendations. Yet explanations have been shown to increase the user's trust in the system in addition to providing other benefits such as scrutability, meaning the ability to verify the validity of recommendations. This gap between accuracy and transparency...

Recommender Systems (RSs) are widely used to help online users discover products, books, news, music, movies, courses, restaurants, etc. Because a traditional recommendation strategy always shows the most relevant items (thus with highest predicted rating), traditional RS’s are expected to make popular items become even more popular and non-popular...

Driven by the explosive growth in available data and decreasing costs of computation, Deep Learning (DL) has found much of its fame in problems involving classification tasks which are considered supervised learning. Deep learning has also been widely used to learn richer and better data representations from big data, without relying too much on hu...

Accurate model-based Collaborative Filtering (CF) approaches, such as Matrix Factorization (MF), tend to be black-box machine learning models that lack interpretability and do not provide a straightforward explanation for their outputs. Yet explanations have been shown to improve the transparency of a recommender system by justifying recommendation...

Post hoc explanations for black-box machine learning models have been criticized for potentially inaccurate surrogate models and computational burden at prediction time. We propose pre hoc and co hoc explainability frameworks that integrate interpretability directly into the training process through an inherently interpretable white-box model. Pre...

This paper presents a local explainability mechanism for robotic grasp failure prediction that enhances machine learning transparency at the instance level. Building upon pre hoc explainability concepts, we develop a neighborhood-based optimization approach that leverages the Jensen–Shannon divergence to ensure fidelity between predictor and explai...

In human–robot collaborative environments, predicting and explaining robotic grasp failures is crucial for effective operation. While machine learning models can predict failures accurately, they often lack transparency, limiting their utility in critical applications. This paper presents a comparative analysis of three post hoc explanation methods...

Despite ongoing efforts to make black-box machine learning models more explainable, transparent, and trustworthy, there is growing advocacy for using only inherently interpretable models for high-stakes decision-making. Post-hoc explanations have been criticized for learning surrogate models that may not accurately reflect the actual mechanisms of...

Detecting and preventing an impending robot grasp failure can prevent object damage during robotic manipulation. When a failure is predicted, a human can opt for teleoperation rather than automation of the grasping task. The operator can also intervene to stop the robot from adjusting, such as tightening the grasp or reducing the speed. In this pap...

Despite ongoing efforts to make black-box machine learning models more explainable, transparent, and trustworthy, there is a growing advocacy for using only inherently interpretable models for high-stake decision making. For instance, post-hoc explanations have recently been criticized because they learn surrogate white-box (explainer) models that,...

Multi-criteria ABC classification is a useful model for automatic inventory management and optimization. This model enables a rapid classification of inventory items into three groups, having varying managerial levels. Several methods, based on different criteria and principles, were proposed to build the ABC classes. However, existing ABC classifi...

The enormous scale of the available information and products on the Internet has necessitated the development of algorithms that intermediate between options and human users. These algorithms attempt to provide the user with relevant information. In doing so, the algorithms may incur potential negative consequences stemming from the need to select...

Bidirectional Transformer architectures are state-of-the-art sequential recommendation models that use a bi-directional representation capacity based on the Cloze task, a.k.a. Masked Language Modeling. The latter aims to predict randomly masked items within the sequence. Because they assume that the true interacted item is the most relevant one, an...

Multi-criteria ABC classification is an effective technique that allows rapid and automatic organization of a growing number of inventory items into classes having different managerial levels. These built classes help decision-makers efficiently control the inventory and optimize the whole supply chain. However, existing ABC classification methods...

Recent research in recommender systems has demonstrated the advantages of pairwise ranking in recommendation. In this work, we focus on the state-of-the-art pairwise ranking loss function, Bayesian Personalized Ranking (BPR), and aim to address two of its limitations, namely: (1) the lack of explainability and (2) exposure bias. We propose a recomm...

User feedback results in different rating patterns due to the users' preferences, cognitive differences, and biases. However, little research has taken into account cognitive biases when building recommender systems. In this paper, we propose novel methods to take into account user polarization into matrix factorization-based recommendation systems...

Recommender systems that rely on Black-box Machine Learning (ML) models generate recommendations without explaining their rationale. However, they are generally more accurate when compared to white-box models, which are transparent and scrutable. One such black-box model is Matrix Factor-ization, a State of the Art recommendation technique that is...

Recent work in recommender systems has emphasized the importance of fairness, with a particular interest in bias and transparency, in addition to predictive accuracy. In this paper, we focus on the state of the art pairwise ranking model, Bayesian Personalized Ranking (BPR), which has previously been found to outperform pointwise models in predicti...

Despite advances in deep learning methods for song recommendation, most existing methods do not take advantage of the sequential nature of song content. In addition, there is a lack of methods that can explain their predictions using the content of recommended songs and only a few approaches can handle the item cold start problem. In this work, we...

This paper shows that least-square estimation (mean calculation) in a reproducing kernel Hilbert space (RKHS) F corresponds to different M-estimators in the original space depending on the kernel function associated with F. In particular, we present a proof of the correspondence of mean estimation in an RKHS for the Gaussian kernel with robust esti...

The closed feedback loop in recommender systems is a common setting that can lead to different types of biases. Several studies have dealt with these biases by designing methods to mitigate their effect on the recommendations. However, most existing studies do not consider the iterative behavior of the system where the closed feedback loop plays a...

The enormous scale of the available information and products on the Internet has necessitated the development of algorithms that intermediate between options and human users. These AI/machine learning algorithms attempt to provide the user with relevant information. In doing so, the algorithms may incur potential negative consequences stemming from...

The closed feedback loop in recommender systems is a common setting that can lead to different types of biases. Several studies have dealt with these biases by designing methods to mitigate their effect on the recommendations. However, most existing studies do not consider the iterative behavior of the system where the closed feedback loop plays a...

Traditionally, machine learning algorithms relied on reliable labels from experts to build predictions. More recently however, algorithms have been receiving data from the general population in the form of labeling, annotations, etc. The result is that algorithms are subject to bias that is born from ingesting unchecked information, such as biased...

The ability to determine whether a robot’s grasp has a high chance of failing, before it actually does, can save significant time and avoid failures by planning for re-grasping or changing the strategy for that special case. Machine Learning (ML) offers one way to learn to predict grasp failure from historic data consisting of a robot’s attempted g...

What we discover and see online, and consequently our opinions and decisions, are becoming increasingly affected by automated machine learned predictions. Similarly, the predictive accuracy of learning machines heavily depends on the feedback data that we provide them. This mutual influence can lead to closed-loop interactions that may cause unknow...

Autoencoders are a common building block of Deep Learning architectures, where they are mainly used for representation learning. They have also been successfully used in Collaborative Filtering (CF) recommender systems to predict missing ratings. Unfortunately, like all black box machine learning models, they are unable to explain their outputs. He...

Autoencoders are a common building block of Deep Learning archi-tectures, where they are mainly used for representation learning.They have also been successfully used in Collaborative Filtering (CF) recommender systems to predict missing ratings. Unfortunately,like all black box machine learning models, they are unable to ex-plain their outputs. He...

Recommender systems are being increasingly used to predict the preferences of users on online platforms and recommend relevant options that help them cope with information overload. In particular, modern model-based collaborative filtering algorithms, such as latent factor models, are considered state-of-the-art in recommendation systems. Unfortuna...

State of the art music recommender systems mainly rely on either Matrix factorization-based collaborative filtering approaches or deep learning architectures. Deep learning models usually use metadata for content-based filtering or predict the next user interaction by learning from temporal sequences of user actions. Despite advances in deep learni...

Early supervised machine learning (ML) algorithms have used reliable labels from experts to build predictions. But recently, these algorithms have been increasingly receiving data from the general population in the form of labels, annotations, etc. The result is that algorithms are subject to bias that is born from ingesting unchecked information,...

Machine Learning (ML) models are increasingly being used in many sectors, ranging from health and education to justice and criminal investigation. Therefore, building a fair and transparent model which conveys the reasoning behind its predictions is of great importance. This chapter discusses the role of explanation mechanisms in building fair mach...

Social networks, along with their “event” organization, planning, and sharing tools, play an important role in connecting and engaging individuals and groups. These online spaces thrive with multifaceted activities and interests which give rise to rich content and user interaction that often crossover to the world of events. For these reasons, the...

Personalized recommender systems are commonly used to filter information in social media, and recommen- dations are derived by training machine learning algorithms on these data. It is thus important to understand how machine learning algorithms, especially recommender systems, behave in polarized environments. We investigate how filtering and disc...

Personalized recommender systems are becoming increasingly relevant and important in the study of polarization and bias, given their widespread use in filtering information spaces. Polarization is a social phenomenon, with serious consequences, in real-life, particularly on social media. Thus it is important to understand how machine learning algor...

The goal of this study is to develop a model that explains the relationship between microRNAs, transcription factors, and their co-target genes. This relationship was previously reported in gene regulatory loops associated with 24 hour (24h) and 7 day (7d) time periods following ischemia-reperfusion injury in a rat’s retina. Using a model system of...

Supporting and opposing loops at 24h and 7d. A total of four sheets included. Sheet names are suffixed with “24h” or “7d” to indicate the IR time point and prefixed with “supporting”, “opposing” to indicate pairs of miRNAs-TFs that are working together or against each other respectively. (XLSX)

Top mediated loops for each class of loops at 24h and 7d. A total of two sheets included for 24h, and 7d respectively. Each sheet contains four additional tables listing the top five mediated loops in each class of mediated loops. (XLSX)

Validated mediated loops 24h and 7d. Partial validation from miRWALK db. A total of six sheets included. Sheet names are suffixed with “24h” or “7d” to indicate the IR time point and prefixed with “MT”, “MM”, or “MTM” to indicate Mediation by TFs, mediation by miRNAs, and mediation by both TFs, and miRNAs respectively. (XLSX)

Mediation result and classification of closed regulatory loops at 24h and 7d. A total of eight sheets included. Sheet names are suffixed with “24h” or “7d” to indicate the IR time point and prefixed with “MT”, “MM”, or “MTM” to indicate Mediation by TFs, mediation by miRNAs, and mediation by both TFs, and miRNAs respectively. Sheets “24h”, and “7d”...

The enormous scale of the available information and products on the Internet has necessitated the development of algorithms that intermediate between options and human users. These algorithms do not select information at random, but attempt to provide the user with relevant information. In doing so, the algorithms may incur potential negative conse...

We describe the construction of a bilingual (English-Russian /Russian-English) semantic network covering basic concepts of computing. To construct the semantic network, we used the Computing Curricular series created during 2000-2015 under the aegis of ACM and IEEE and the current standards of IT specialists training in Russia, as well as some othe...

We describe a service-based approach that provides a natural language interface to legacy information systems, built on top of relational database management systems. The long term goal is to make data management and analysis accessible to a wider range of users for a diverse range of purposes and to simplify the decision making process. We present...

Topic Models are statistical models that can be used for discovering the abstract “topics” that may occur in a text corpus, however they face dramatic challenges when coping with very sparse and yet topically diverse micro-blog posts such as tweets. In such streams, not only are the topics very diverse, but also the vocabulary is huge, making the s...

We bring to the fore of the recommender system research community, an inconvenient truth about the current state of understanding how recommender system algorithms and humans influence one another, both computationally and cognitively. Unlike the great variety of supervised machine learning algorithms which traditionally rely on expert input labels...

Early supervised machine learning algorithms have relied on reliable expert labels to build predictive models. However, the gates of data generation have recently been opened to a wider base of users who started participating increasingly with casual labeling, rating, annotating, etc. The increased online presence and participation of humans has le...

Online third party marketplaces link buyers and sellers by providing a neutral platform for exchange. However, this requires buyers to assess the quality of goods without being able to handle or sample them. Recent research has proposed extending the warranting principle, an emerging theory of online interpersonal impression formation, to the judge...

Abstract Refining city services is gradually being placed in the hands of the citizens, or, as in the case of IBM’s initiative, Blet’s build a planet of smarter cities^ (https:// www-03.ibm.com/press/us/en/pressrelease/35573.wss), at their fingertips. By reducing cost and gaining control in building smart transportation management systems, IBM prov...

Background The volume of biomedical literature and its underlying knowledge base is rapidly expanding, making it beyond the ability of a single human being to read through all the literature. Several automated methods have been developed to help make sense of this dilemma. The present study reports on the results of a text mining approach to extrac...

Explanations have been shown to increase the user's trust in recommendations in addition to providing other benefits such as scrutability, which is the ability to verify the validity of recommendations. Most explanation methods are designed for classical neighborhood-based Collaborative Filtering (CF) or rule-based methods. For the state of the art...

We present a novel environment for knowledge modeling and visualization for domain ontology design based on metaknowledge representation (metalevel) that was also implemented within an ontological paradigm framework. Metaknowledge representation makes the visual ontology editor more adaptable to the user's individual preferences. This improved adap...

Clustering algorithms and tips for success: choosing the right algorithm, distance measure, handling initialization, cluster result evaluation methods and some tips for python implementation. The last slide has a Clustering recipe concept map! Check it out!

We describe a proposed universal design infrastructure that aims at promoting better opportunities for students with disabilities in STEM programs to understand multimedia teaching material. The Accessible Educational STEM Videos Project aims to transform learning and teaching for students with disabilities through integrating synchronized captione...

We demonstrate a new deep learning autoencoder network, trained by a nonnegativity constraint algorithm (NCAE), that learns features which show part-based representation of data. The learning algorithm is based on constraining negative weights. The performance of the algorithm is assessed based on decomposing data into parts and its prediction perf...

From e-commerce to e-learning, recommendation systems have given birth to an important and thriving research niche and have been deployed in a variety of application areas over the last decade. In particular, in the technology-enhanced learning (TEL) field, recommendation systems have attracted increasing interest, especially with the rise of educa...

In this paper, we describe our solution to the RecSys2014 challenge and results on the test set. We briefly describe some of the challenges, then describe the methodology which starts with feature extraction and construction using the provided tweet data, in combination with IMDB as an external source. Feature construction also involved computing s...

In this paper, we describe our solution to the RecSys2014 challenge and results on the test set. We briefly describe some of the challenges, then describe the methodology which starts with feature extraction and construction using the provided tweet data, in combination with IMDB as an external source. Feature construction also involved computing s...

Social media has recently emerged as an invaluable source of information for decision making. Social media information reflects the interests of virtual communities in a spontaneous and timely manner. The need to understand the massive streams of data generated by social media platforms, such as Twitter and Facebook, has motivated researchers to us...

Although clustering is an unsupervised learning approach, most clustering algorithms require the setting of parameters (such as the number of clusters, minimum density or distance threshold) in advance to work properly. Moreover, discovering an appropriate set of clusters is a difficult task since clusters can have any shape, size and density and i...

We present a cross-modal recommendation engine that lever- ages multiple domains of data while performing matrix fac- torization. We show how our approach has the potential to alleviate the cold-start problem for new items, one of the notorious limitations of Collaborative Filtering (CF) tech- niques.

Continuous social text streams, such as tweets, provide a timeline of discussions. Topic modeling techniques such as Latent Dirichlet Allocation (LDA) have been used to extract the topics being discussed on social media streams. Recently, Online LDA has been proposed as a fast alternative for topic extraction, based on on-line stochastic optimizati...

We describe the methodology that we followed to automat-ically extract topics corresponding to known events pro-vided by the SNOW 2014 challenge in the context of the SocialSensor project. A data crawling tool and selected fil-tering terms were provided to all the teams. The crawled data was to be divided in 96 (15-minute) timeslots spanning a 24 h...

Kernel clustering methods have been used successfully to cluster non linearly separable data. In this paper, we propose a modification of the Kernel K-means, called the Multi-Scale Kernel K-means, that addresses one important challenge, which is the automated estimation of the kernel scale parameters for data containing clusters with different scal...

In this paper, we describe a fully automatic learner modeling approach in learning management systems, taking into account learners' educational preferences including learning styles. We propose a learner model with three components: the learner's profile, learner's knowledge, and learner's educational preferences. The learner's profile represents...

We propose a Seed-based Inter-Domain Supervised (IDS) framework to handle possibly diverse data formats, mixed-type attributes and different sources of data. This approach can be used for combining diverse representations of the data, in particular where data comes from different sources, some of which may be unreliable or uncertain, or for exploit...

Although clustering is an unsupervised learning approach, most clustering algorithms require setting several parameters (such as the number of clusters, minimum density or distance threshold) in advance to work properly. In this paper, we eliminate the necessity of setting the minimum cluster size parameter of the Randomized Gravitational Clusterin...

Recent years have seen an increasing interest in clustering data comprising multiple domains or modalities, such as categorical, numerical and transactional, etc. This kind of data is sometimes found within the context of clustering multiview, heterogeneous, or multimodal data. Traditionally, different types of attributes or domains have been handl...

We propose a new methodology for clustering data comprising multiple domains or parts, in such a way that the separate domains mutually supervise each other within a semi-supervised learning framework. Unlike existing uses of semi-supervised learning, our methodology does not assume the presence of labels from part of the data, but rather, each of...

We propose a Genetic algorithm for document clustering, where an evolutionary multimodal optimization algorithm evolves candidate cluster representative solutions to search for dense regions in the sparse high dimensional vector space of text documents. The evolution affects not only the document cluster representatives but also the genetic operato...

This paper evaluates the robustness of two types of unsupervised learning methods, which work in feature spaces induced by a kernel function, kernel k-means and kernel symmetric non-negative matrix factorization. The main hypothesis is that the use of non-linear kernels makes these clustering algorithms more robust to noise and outliers. The hypoth...

Clustering data streams is a challenging problem that has received significant attention in the recent decade. In this paper, we address the hitherto inadequately addressed challenge of managing the output of stream clustering. This task comprises the continuous cluster model validation, monitoring, trend and change detection, and summarization of...