arXiv subjects

Talel Abdessalem

Publications and source records attributed to Talel Abdessalem.

12 recordsLinked to original sources

arXiv2024

Explaining the (Not So) Obvious: Simple and Fast Explanation of STAN, a Next Point of Interest Recommendation System

A lot of effort in recent years have been expended to explain machine learning systems. However, some machine learning methods are inherently explainable, and thus are not completely black box. This enables the developers to make sense of the output without a developing a complex and expensive explainability technique. Besides that, explainability should be tailored to suit the context of the problem. In a recommendation system which relies on collaborative filtering, the recommendation is based on the behaviors of similar users, therefore the explanation should tell which other users are similar to the current user. Similarly, if the recommendation system is based on sequence prediction, the explanation should also tell which input timesteps are the most influential. We demonstrate this philosophy/paradigm in STAN (Spatio-Temporal Attention Network for Next Location Recommendation), a next Point of Interest recommendation system based on collaborative filtering and sequence prediction. We also show that the explanation helps to "debug" the output.

cs.IR↗

arXiv2021

Exact and Approximate Algorithms for Computing Betweenness Centrality in Directed Graphs

Graphs (networks) are an important tool to model data in different domains. Real-world graphs are usually directed, where the edges have a direction and they are not symmetric. Betweenness centrality is an important index widely used to analyze networks. In this paper, first given a directed network $G$ and a vertex $r \in V(G)$, we propose an exact algorithm to compute betweenness score of $r$. Our algorithm pre-computes a set $\mathcal{RV}(r)$, which is used to prune a huge amount of computations that do not contribute to the betweenness score of $r$. Time complexity of our algorithm depends on $|\mathcal{RV}(r)|$ and it is respectively $Θ(|\mathcal{RV}(r)|\cdot|E(G)|)$ and $Θ(|\mathcal{RV}(r)|\cdot|E(G)|+|\mathcal{RV}(r)|\cdot|V(G)|\log |V(G)|)$ for unweighted graphs and weighted graphs with positive weights. $|\mathcal{RV}(r)|$ is bounded from above by $|V(G)|-1$ and in most cases, it is a small constant. Then, for the cases where $\mathcal{RV}(r)$ is large, we present a simple randomized algorithm that samples from $\mathcal{RV}(r)$ and performs computations for only the sampled elements. We show that this algorithm provides an $(ε,δ)$-approximation to the betweenness score of $r$. Finally, we perform extensive experiments over several real-world datasets from different domains for several randomly chosen vertices as well as for the vertices with the highest betweenness scores. Our experiments reveal that for estimating betweenness score of a single vertex, our algorithm significantly outperforms the most efficient existing randomized algorithms, in terms of both running time and accuracy. Our experiments also reveal that our algorithm improves the existing algorithms when someone is interested in computing betweenness values of the vertices in a set whose cardinality is very small.

cs.DS↗

arXiv2020

River: machine learning for streaming data in Python

River is a machine learning library for dynamic data streams and continual learning. It provides multiple state-of-the-art learning methods, data generators/transformers, performance metrics and evaluators for different stream learning problems. It is the result from the merger of the two most popular packages for stream learning in Python: Creme and scikit-multiflow. River introduces a revamped architecture based on the lessons learnt from the seminal packages. River's ambition is to be the go-to library for doing machine learning on streaming data. Additionally, this open source package brings under the same umbrella a large community of practitioners and researchers. The source code is available at https://github.com/online-ml/river.

cs.LG↗

arXiv2020

Adaptive XGBoost for Evolving Data Streams

Boosting is an ensemble method that combines base models in a sequential manner to achieve high predictive accuracy. A popular learning algorithm based on this ensemble method is eXtreme Gradient Boosting (XGB). We present an adaptation of XGB for classification of evolving data streams. In this setting, new data arrives over time and the relationship between the class and the features may change in the process, thus exhibiting concept drift. The proposed method creates new members of the ensemble from mini-batches of data as new data becomes available. The maximum ensemble size is fixed, but learning does not stop when this size is reached because the ensemble is updated on new data to ensure consistency with the current concept. We also explore the use of concept drift detection to trigger a mechanism to update the ensemble. We test our method on real and synthetic data with concept drift and compare it against batch-incremental and instance-incremental classification methods for data streams.

cs.LG↗

arXiv2018

Novel Adaptive Algorithms for Estimating Betweenness, Coverage and k-path Centralities

An important index widely used to analyze social and information networks is betweenness centrality. In this paper, first given a directed network $G$ and a vertex $r\in V(G)$, we present a novel adaptive algorithm for estimating betweenness score of $r$. Our algorithm first computes two subsets of the vertex set of $G$, called $\mathcal{RF}(r)$ and $\mathcal{RT}(r)$, that define the sample spaces of the start-points and the end-points of the samples. Then, it adaptively samples from $\mathcal{RF}(r)$ and $\mathcal{RT}(r)$ and stops as soon as some condition is satisfied. The stopping condition depends on the samples met so far, $|\mathcal{RF}(r)|$ and $|\mathcal{RT}(r)|$. We show that compared to the well-known existing methods, our algorithm gives a more efficient $(λ,δ)$-approximation. Then, we propose a novel algorithm for estimating $k$-path centrality of $r$. Our algorithm is based on computing two sets $\mathcal{RF}(r)$ and $\mathcal{D}(r)$. While $\mathcal{RF}(r)$ defines the sample space of the source vertices of the sampled paths, $\mathcal{D}(r)$ defines the sample space of the other vertices of the paths. We show that in order to give a $(λ,δ)$-approximation of the $k$-path score of $r$, our algorithm requires considerably less samples. Moreover, it processes each sample faster and with less memory. Finally, we empirically evaluate our proposed algorithms and show their superior performance. Also, we show that they can be used to efficiently compute centrality scores of a set of vertices.

cs.DS↗

arXiv2018

Scikit-Multiflow: A Multi-output Streaming Framework

Scikit-multiflow is a multi-output/multi-label and stream data mining framework for the Python programming language. Conceived to serve as a platform to encourage democratization of stream learning research, it provides multiple state of the art methods for stream learning, stream generators and evaluators. scikit-multiflow builds upon popular open source frameworks including scikit-learn, MOA and MEKA. Development follows the FOSS principles and quality is enforced by complying with PEP8 guidelines and using continuous integration and automatic testing. The source code is publicly available at https://github.com/scikit-multiflow/scikit-multiflow.

cs.LG↗

arXiv2018

Discriminative Distance-Based Network Indices with Application to Link Prediction

In large networks, using the length of shortest paths as the distance measure has shortcomings. A well-studied shortcoming is that extending it to disconnected graphs and directed graphs is controversial. The second shortcoming is that a huge number of vertices may have exactly the same score. The third shortcoming is that in many applications, the distance between two vertices not only depends on the length of shortest paths, but also on the number of shortest paths. In this paper, first we develop a new distance measure between vertices of a graph that yields discriminative distance-based centrality indices. This measure is proportional to the length of shortest paths and inversely proportional to the number of shortest paths. We present algorithms for exact computation of the proposed discriminative indices. Second, we develop randomized algorithms that precisely estimate average discriminative path length and average discriminative eccentricity and show that they give $(ε,δ)$-approximations of these indices. Third, we perform extensive experiments over several real-world networks from different domains. In our experiments, we first show that compared to the traditional indices, discriminative indices have usually much more discriminability. Then, we show that our randomized algorithms can very precisely estimate average discriminative path length and average discriminative eccentricity, using only few samples. Then, we show that real-world networks have usually a tiny average discriminative path length, bounded by a constant (e.g., 2). Fourth, in order to better motivate the usefulness of our proposed distance measure, we present a novel link prediction method, that uses discriminative distance to decide which vertices are more likely to form a link in future, and show its superior performance compared to the well-known existing measures.

cs.DS↗

arXiv2017

Metropolis-Hastings Algorithms for Estimating Betweenness Centrality in Large Networks

Betweenness centrality is an important index widely used in different domains such as social networks, traffic networks and the world wide web. However, even for mid-size networks that have only a few hundreds thousands vertices, it is computationally expensive to compute exact betweenness scores. Therefore in recent years, several approximate algorithms have been developed. In this paper, first given a network $G$ and a vertex $r \in V(G)$, we propose a Metropolis-Hastings MCMC algorithm that samples from the space $V(G)$ and estimates betweenness score of $r$. The stationary distribution of our MCMC sampler is the optimal sampling proposed for betweenness centrality estimation. We show that our MCMC sampler provides an $(ε,δ)$-approximation, where the number of required samples depends on the position of $r$ in $G$ and in many cases, it is a constant. Then, given a network $G$ and a set $R \subset V(G)$, we present a Metropolis-Hastings MCMC sampler that samples from the joint space $R$ and $V(G)$ and estimates relative betweenness scores of the vertices in $R$. We show that for any pair $r_i, r_j \in R$, the ratio of the expected values of the estimated relative betweenness scores of $r_i$ and $r_j$ respect to each other is equal to the ratio of their betweenness scores. We also show that our joint-space MCMC sampler provides an $(ε,δ)$-approximation of the relative betweenness score of $r_i$ respect to $r_j$, where the number of required samples depends on the position of $r_j$ in $G$ and in many cases, it is a constant.

cs.DS↗

arXiv2017

Pessimistic Uplift Modeling

Uplift modeling is a machine learning technique that aims to model treatment effects heterogeneity. It has been used in business and health sectors to predict the effect of a specific action on a given individual. Despite its advantages, uplift models show high sensitivity to noise and disturbance, which leads to unreliable results. In this paper we show different approaches to address the problem of uplift modeling, we demonstrate how disturbance in data can affect uplift measurement. We propose a new approach, we call it Pessimistic Uplift Modeling, that minimizes disturbance effects. We compared our approach with the existing uplift methods, on simulated and real data-sets. The experiments show that our approach outperforms the existing approaches, especially in the case of high noise data environment.

cs.LG↗

arXiv2017

Social media marketing analytics : a multicultural approach applied to the beauty \& cosmetic sector

We present in this paper a multicultural approach to social media marketing analytics, applied in two Facebook brand pages: French (individualistic culture, the country home of the brand) versus Saudi Arabian (collectivistic culture, one of its country hosts), which are published by an internationalbeauty \& cosmetics firm. Using social network analysis and content analysis, we identify the most popular posts and the most influential users within these two brand pages and highlight the different communities emerging from brand and users interactions. These communities seem to be culture oriented when they are constructed around socialization branded posts and product-line oriented when advertising branded posts are concerned.

cs.SI↗

arXiv2013

Improving tag recommendation by folding in more consistency

Tag recommendation is a major aspect of collaborative tagging systems. It aims to recommend tags to a user for tagging an item. In this paper we present a part of our work in progress which is a novel improvement of recommendations by re-ranking the output of a tag recommender. We mine association rules between candidates tags in order to determine a more consistent list of tags to recommend. Our method is an add-on one which leads to better recommendations as we show in this paper. It is easily parallelizable and morever it may be applied to a lot of tag recommenders. The experiments we did on five datasets with two kinds of tag recommender demonstrated the efficiency of our method.

cs.IR↗

arXiv2012

Dynamic recommender system : using cluster-based biases to improve the accuracy of the predictions

It is today accepted that matrix factorization models allow a high quality of rating prediction in recommender systems. However, a major drawback of matrix factorization is its static nature that results in a progressive declining of the accuracy of the predictions after each factorization. This is due to the fact that the new obtained ratings are not taken into account until a new factorization is computed, which can not be done very often because of the high cost of matrix factorization. In this paper, aiming at improving the accuracy of recommender systems, we propose a cluster-based matrix factorization technique that enables online integration of new ratings. Thus, we significantly enhance the obtained predictions between two matrix factorizations. We use finer-grained user biases by clustering similar items into groups, and allocating in these groups a bias to each user. The experiments we did on large datasets demonstrated the efficiency of our approach.

cs.LG↗