Associate Professor (Docent)
Artificial Intelligence and Data Science
Skövde University - School of informatics, Skövde, Sweden
- Federated Learning / Decentralized AI
- Data Science & Artificial Intelligence
- Machine Learning / DL / NLP / Conv AI
- Data Mining / Information Retrieval
About Me!
Dr. Addi Ait-Mlouk is an Associate Professor (Docent) in Artificial Intelligence and Data Science at the School of Informatics,
University of Skövde, Sweden. He served as the Director of the Data Science Master’s Programme (2023–2024), showcasing his leadership
in shaping AI education and mentoring the next generation of data scientists.
Before joining the University of Skövde, Addi was a Postdoctoral Researcher at Uppsala University (2020–2022), where he worked on a
Scalable Federated Learning project at the Department of Information Technology, Division of Scientific Computing. Prior to that, he was a
Postdoctoral Researcher at Umeå University (2018–2020), contributing to a WASP project on Privacy-Aware Data Federation for heterogeneous registry data.
His research focuses on solving real-world challenges through cutting-edge AI and data science techniques. His expertise spans Federated Learning, Machine
Learning, Data Mining, Information Retrieval, Conversational AI, Open Linked Data, Knowledge Graphs, and Natural Language Processing.
Addi holds a Master’s degree in Information Systems Engineering and a PhD in Data Mining from Cadi Ayyad University, Marrakech, earned in 2011 and 2018, respectively.
Beyond academia, he is a highly sought-after consultant and trainer specializing in CRM, ERP, Data Science, and AI. Passionate about emerging technologies, he collaborates
with businesses to drive digital transformation and innovation.
He actively contributes to the scientific community as a reviewer and program committee member for leading conferences and journals, including ICDM, ECIR, and ECML-PKDD.
Born in Imi N’Ouakka, Morocco, in 1990, Addi is always open to new collaboration opportunities in academia and industry. Feel free to connect with him at aitmlouk@gmail.com.
Academic Positions
-
Present 09/2022
-
09/2022 09/2020
Postdoc Researcher
Scalable Federated Machine Learning
Uppsala University - Computer science, Uppsala, Sweden
-
09/2020 09/2018
Postdoc Researcher
Privacy-aware Federated Database Infrastructure
Umeå University - Computer science, Umeå, Sweden
-
07/2018 09/2017
Assistant Professor
Private University Marrakech - Computer science, Marrakech, Morocco
-
06/2017 02/2015
Teaching Assistant
Cadi Ayyad University, Faculty of Science Semlalia, Marrakech, Morocco
Education & Training
-
Ph.D. 2014-2018
Ph.D. in Computer Science [ Data Mining ]
Cadi Ayyad University, Faculty of Science Semlalia, Marrakech, Morocco
-
M.S. 2011-2013
Master of Information Systems Engineering
Cadi Ayyad University, Faculty of Science Semlalia, Marrakech, Morocco
-
B.Sc. 2008-2011
B.Sc. in Software Engineering
Mly Ismail University, Sciences and Technologies Faculty, Errachidia, Morocco
Services
- AI & Machine Learning Consulting – Helping businesses and researchers integrate AI solutions.
- Federated Learning & Decentralized AI – Implementing privacy-preserving AI solutions.
- Data Science & Analytics – Analyzing data for insights and decision-making.
- ERP & AI – Developing tailored ERP and AI solutions for businesses.
- Educational Training – Conducting AI, Data Science, and ERP-related training.
Research Summary
Research in an academic environment provides a unique opportunity to solve real-life problems and vision the technology of the future. My current research interests are concerned with the development of methods and algorithms for decentralized artificial intelligence, machine learning, and data mining. Computer-generated data is becoming ubiquitous in all real-life applications ranging from business to more serious applications like medical and scientific data visualization. In the past two decades, there have been many advances in the field of machine learning and AI. Despite numerous technological advances in both graphics hardware as well as algorithms, many challenging problems still limit the realism and effectiveness of computer-generated data.
My main research goal is to create tools and algorithms for efficient machine learning and AI with private data. Nowadays, data regulated under GDPR and HIPPA are becoming more and more complex to analyze, and as the Web continues to grow, it is important to understand these huge amounts of data for two reasons, first, to better serve customers. For example, giant companies use the data to suggest products, and movies to their users; Google and similar companies are providing personalized search results, targeted advertising, and many other services. Second, an in-depth understanding of the data allows us to leverage it for a variety of purposes. By preserving data privacy, I am particularly interested in developing new methods and algorithms for knowledge discovery, machine learning, AI applications and present them as a valid element for decision-makers.
While pursuing my Ph.D., I implemented state-of-the-art knowledge discovery techniques, big data, machine learning, and multiple criteria decision analysis to find solutions for problems around us such as road safety. I also proposed methodologies to evaluate association rules issued from the KDD process, which has been used in real-life challenges, for instance, road safety, text mining, and social networks. Consequently, I had proposed many approaches to process and analyze large datasets. Besides, I have a substantial amount of industry experience in data mining-related research and applied machine learning. From my previous experiences and current interests, I see myself doing research in data mining, social science, transportation, and business, with a special focus on data mining, Artificial Intelligence, and machine learning.
In the upcoming years, I would like to explore a broad range of topics both within machine learning and federated learning areas that improve efficiency, quality, and effectiveness of depiction while exploiting fields ranging from computer science to medical and business. My research agenda decomposes into many conceptual problems, which will often require technical innovations along three dimensions: models, algorithms, and experimental evaluation. Indeed, I see myself doing research in federated learning with researchers from computer science and other disciplines such as healthcare and business. Following are a few suggested projects but not limited to:
Interests
- Artificial Intelligence
- Federated Learning
- Decentralized AI
- Machine Learning/DL
- Natural Language Processing
- Large Language Models (LLMs)
- Conversational AI
- Semantic web/ Knowledge Graph
- Data Mining
- Multiple Criteria Decision Analysis (MCDA)
Ph.D Students
In Progress
Integrated modelling for sustainable and optimized steel manufacturing processes
Fatiha Ait Baali
Federated Fine-Tuning of LLMs for Domain-Specific Applications: Balancing Personalization and Privacy.
Chaima Lhasnaoui
Enabling Federated Learning on the Edge for Enhanced Privacy in Medical Imaging.
Research Projects
-
ProcTwin aims to develop a demonstration platform to predict and visualize best use of multiple processing steps in a steel manufacturing chain. The methodology includes intelligent coupling of interconnected processing steps by numerical simulation, soft sensors, process data and distributed machine learning. Integrated numerical modelling that captures the interactions, relations, and feedback loops between various processing stations enables prediction for smart optimization of energy efficiency and product quality in the steel manufacturing.
The main idea is that instead of optimising each subsequent process component’s AI model separately, there is a benefit to have a distributed AI model that learns how to share information between such local AI models, to optimise the overall goal. We apply this in a chain of processes for a certain production output quality. The concept is meant to be used in steel manufacturing production lines, but the general case can be applied wherever there is a need for distributed AI with interacting and collaborating components that need to share knowledge for a common goal. The challenge is to learn what model knowledge to exchange, and how, to reach the common goal. -
Motivated by the need for AI on private data and facilitating collaboratve machine learning. FEDn is an open-source, modular and ML-framework agnostic framework for Federated Machine Learning (FedML) developed by researchers from Uppsala university and Scaleout Systems. FEDn enables highly scalable cross-silo and cross-device use-cases over FEDn networks.
-
Motivated by the need for cross-database analysis on heterogeneous data and facilitating distributed register data usage for research purposes in Sweden, Umeå University is investing in research and development of expertise to build an infrastructure of federated database integrating major research data sources at the university and several national databases. To achieve this goal, a group for register-based research from different departments of Umeå University are connecting for this project "Privacy-aware Data Federation Infrastructure on Heterogeneous Register Data". The major themes of this project are conducting academic research on data federation techniques and privacy preservation on register data sharing. A data federation infrastructure will be thus constructed for practical usage and research testbed. This white paper presents the project background, technology challenges, overall framework, and implementation solutions
-
In this project, we are interested in the problem of extracting association rules by introducing new algorithms and approaches. In general, an association rule is a conditional implication between sets of
Teaching Summary
My main teaching areas of interest center around data mining, data science, big data, machine learning, AI, NLP, cloud computing, semantic web, fuzzy logic, business information system, ERPs, and multiple criteria decision analysis. I enjoy teaching because of the opportunity to connect with people and share my love of computer science. Each teaching setting has challenged me to expand my definition of teacher. I have pursued extensive teaching in many universities among them Cadi Ayyad University, Private University of Marrakech, Umeå University, and Uppsala University. For all of these courses, I arrange research and scholarly activities to teach students about web development, data analytics, data science, machine learning, databases, and business information systems. I also assign projects throughout the semester and organize weekly or bimonthly meetings in which we discuss the progress of these activities. These additional responsibilities ensure that being a member of my research team is an educational experience, and help to get students involved in the work we are doing. Student performance on assignments and examinations also provides benchmarks for examining student progress, both within and across semesters.
In the context of online teaching, I have pursued extensive mentoring and supervising students at Openclassrooms. I have mentored more than 300 graduate students in a different area of computer science ranging from web development to artificial intelligence.
My major goal is to teach students how to apply general principles to novel settings, many of my examination questions ask them to apply concepts learned in class to new contexts. In some of my undergraduate classes, however, I found that many students had difficulty with the applied questions. Thus, I now teach students how to generalize their knowledge through in-class activities designed to explicitly model this type of thinking, and I have subsequently observed important improvement in student performance on applied questions. In the future, I would like to continue my relationship with the teaching service office and taking advantage of local resources. I would like to integrate ideas I've learned through recent training and experiences in coordination with student feedback to come up with innovative strategies for teaching computer science and make it easy for the student on both theoretical and industry aspects.
To summarise, my careful observations of the teaching methods and philosophies that I experienced throughout my education provide a valuable collection of effective teaching techniques from which I draw. These various approaches lead me to adhere to a principle of clarity that demands that my lectures, expectations of students, and educational goals for them be as clear as possible in the interest of maximizing their educations. Indeed, I see myself teaching various courses ranging from data structure to AI and business information systems. Following are some topic I taught but not limited to:
Courses
- Applied Cloud Compution (7.5 credits)
- Data Engineering II (7.5 credits)
- Data Mining (7.5 credits)
- Intro to Data science (15 credits)
- Object Oriented Programming (7.5 credits)
- Project in Data Science A1F (15 credits)
- Degree project in IT with a focus on Data Science A1E (15 credits)
- Degree project in IT with a focus on Data Science A2E (15 credits)
- Database administration (DBA)
- Web development (PhP, JS, HTML)
- Programming C/C++
- Machine Learning (ML)
- Enterprise Resources Planning (ERP)
Pedagogical Courses (Educational Development)
- 11/2021 — 12/2021: Curious about leadership - for future research group leaders, Uppsala University, Uppsala, Sweden.
- 11/2021 — 11/2021: Supervising student presentations in theory and in practice, Uppsala University, Uppsala, Sweden.
- 08/2021 — 10/2021: Academic Teacher Training Course 1, Uppsala University, Uppsala, Sweden.
- 02/2020 — 04/2021: Supervising Doctoral Students, spring 2021, Uppsala University, Uppsala, Sweden.
- 04/2020 — 05/2021: Assessment, grading and feedback, Uppsala University, Uppsala, Sweden.
- 03/2020 — 05/2021: Supervising Students for Degree Projects, Uppsala University, Uppsala, Sweden.
Research Projects
-
ProcTwin aims to develop a demonstration platform to predict and visualize best use of multiple processing steps in a steel manufacturing chain. The methodology includes intelligent coupling of interconnected processing steps by numerical simulation, soft sensors, process data and distributed machine learning. Integrated numerical modelling that captures the interactions, relations, and feedback loops between various processing stations enables prediction for smart optimization of energy efficiency and product quality in the steel manufacturing.
The main idea is that instead of optimising each subsequent process component’s AI model separately, there is a benefit to have a distributed AI model that learns how to share information between such local AI models, to optimise the overall goal. We apply this in a chain of processes for a certain production output quality. The concept is meant to be used in steel manufacturing production lines, but the general case can be applied wherever there is a need for distributed AI with interacting and collaborating components that need to share knowledge for a common goal. The challenge is to learn what model knowledge to exchange, and how, to reach the common goal. -
Motivated by the need for AI on private data and facilitating collaboratve machine learning. FEDn is an open-source, modular and ML-framework agnostic framework for Federated Machine Learning (FedML) developed by researchers from Uppsala university and Scaleout Systems. FEDn enables highly scalable cross-silo and cross-device use-cases over FEDn networks.
-
Motivated by the need for cross-database analysis on heterogeneous data and facilitating distributed register data usage for research purposes in Sweden, Umeå University is investing in research and development of expertise to build an infrastructure of federated database integrating major research data sources at the university and several national databases. To achieve this goal, a group for register-based research from different departments of Umeå University are connecting for this project "Privacy-aware Data Federation Infrastructure on Heterogeneous Register Data". The major themes of this project are conducting academic research on data federation techniques and privacy preservation on register data sharing. A data federation infrastructure will be thus constructed for practical usage and research testbed. This white paper presents the project background, technology challenges, overall framework, and implementation solutions
-
In this project, we are interested in the problem of extracting association rules by introducing new algorithms and approaches. In general, an association rule is a conditional implication between sets of binary attributes called items. The extraction of such rules is composed of two main steps which are the extraction of frequent itemsets and the generation of association rules from them. The complexity of each of these steps is exponential : the number of frequent itemsets is exponential, and the number of association rules extracted can be very high, due to the quality measures used. In the literature, the extraction of the association rules is composed into two main difficulties, the response time and the memory space. To overcome these difficulties, we propose in this thesis three main contributions respectively allowing the extraction of relevant association rules, the integration of the spatial component into the extraction process, and mining relevant association rules from big data. In the first contribution, we propose an extraction approach of the relevant association rules based on multicriteria decision analysis. Then, in the second contribution, we propose an efficient algorithm for extracting spatial predicates from which frequent sets of items and spatial association rules can be generated based on the preparation of the spatial context and the fuzzy set theory. We also proposed in the third contribution a distributed algorithm for the extraction of association rules from Big Data. Using these contributions, we were able to extract the relevant association rules and reduce the execution time and the memory space. In addition, in order to test concretely the contribution of the proposed solutions, we designed and developed a software prototype consisting of three interfaces. The first entitled ARM interface, is an interactive web interface dedicated to the extraction of association rules. The second interface, entitled MCDA interface, it is an interactive web interface dedicated to the evaluation and extraction of relevant association rules. For the last one, entitled Time Series Forcasting, it is an interactive web interface dedicated to the prediction of road accidents. Moreover, interactive and user-friendly interfaces have been developed by using R language and rshiny. Finally, the experiments conducted on some databases on road accidents in Morocco show the significant feasibility of our contributions.
Selected Publications
I have collaborated actively with researchers in several disciplines of computer science, particularly data mining, transportation, multiple criteria analysis, smart grid, and VANETs. I have published many articles in prestigious journals and conference papers and I have served on roughly thirty conference and workshop program committees and served as the reviewer for many journals and conferences.
Filter by type:
Sort by year:
Fouille de données et analyse de qualité des règles d’association dans les bases de données massives : Application dans le domaine de la sécurité routière.
Ph.D Thesis HAL, 2018
Abstract
In this thesis, we are interested in the problem of extracting association rules by introducing new algorithms and approaches. In general, an association rule is a conditional implication between sets of binary attributes called items. The extraction of such rules is composed of two main steps which are the extraction of frequent itemsets and the generation of association rules from them. The complexity of each of these steps is exponential : the number of frequent itemsets is exponential, and the number of association rules extracted can be very high, due to the quality measures used. In the literature, the extraction of the association rules is composed into two main difficulties, the response time and the memory space. To overcome these difficulties, we propose in this thesis three main contributions respectively allowing the extraction of relevant association rules, the integration of the spatial component into the extraction process, and mining relevant association rules from big data. In the first contribution, we propose an extraction approach of the relevant association rules based on multicriteria decision analysis. Then, in the second contribution, we propose an efficient algorithm for extracting spatial predicates from which frequent sets of items and spatial association rules can be generated based on the preparation of the spatial context and the fuzzy set theory. We also proposed in the third contribution a distributed algorithm for the extraction of association rules from Big Data. Using these contributions, we were able to extract the relevant association rules and reduce the execution time and the memory space. In addition, in order to test concretely the contribution of the proposed solutions, we designed and developed a software prototype consisting of three interfaces. The first entitled ARM interface, is an interactive web interface dedicated to the extraction of association rules. The second interface, entitled MCDA interface, it is an interactive web interface dedicated to the evaluation and extraction of relevant association rules. For the last one, entitled Time Series Forcasting, it is an interactive web interface dedicated to the prediction of road accidents. Moreover, interactive and user-friendly interfaces have been developed by using R language and rshiny. Finally, the experiments conducted on some databases on road accidents in Morocco show the significant feasibility of our contributions.
Abstract
Federated machine learning has great promise to overcome the input privacy challenge in machine learning. The appearance of several projects capable of simulating federated learning has led to a corresponding rapid progress on algorithmic aspects of the problem. However, there is still a lack of federated machine learning frameworks that focus on fundamental aspects such as scalability, robustness, security, and performance in a geographically distributed setting. To bridge this gap we have designed and developed the FEDn framework. A main feature of FEDn is to support both cross-device and cross-silo training settings. This makes FEDn a powerful tool for researching a wide range of machine learning applications in a realistic setting.
Fuzzy set theory-based approach for mining spatial association rules: road accident as a case study. International Journal of Geo-Information.
Journal Submitted International Journal of Geo-Information
Abstract
TBA......
Some Algorithms for Solving Neutrosophic Shortest Path Problem. The International Journal of Research on Intelligent Systems for Real Life Complex Problems
Journal[Accepted], 2020
Abstract
TBA......
WINFRA: A Web-Based Platform for Semantic Data Retrieval and Data Analytics
Journal Mathematics 2020, https://doi.org/10.3390/math8112090
Abstract
Given the huge amount of heterogeneous data stored in different locations, it needs to be federated and semantically interconnected for further use. This paper introduces WINFRA, a comprehensive open-access platform for semantic web data and advanced analytics based on natural language processing (NLP) and data mining techniques (e.g., association rules, clustering, classification based on associations). The system is designed to facilitate federated data analysis, knowledge discovery, information retrieval, and new techniques to deal with semantic web and knowledge graph representation. The processing step integrates data from multiple sources virtually by creating virtual databases. Afterwards, the developed RDF Generator is built to generate RDF files for different data sources, together with SPARQL queries, to support semantic data search and knowledge graph representation. Furthermore, some application cases are provided to demonstrate how it facilitates advanced data analytics over semantic data and showcase our proposed approach toward semantic association rules.
kbot: a knowledge graph based chatbot for natural language understanding over linked data
Journal in IEEE Access, vol. 8, pp. 149220-149230, 2020, doi: 10.1109/ACCESS.2020.3016142..
Abstract
With the rapid progress of the semantic web, a huge amount of structured data has become available on the web in the form of knowledge bases (KBs). Making these data accessible and useful for end-users is one of the main objectives of chatbots over linked data. Building a chatbot over linked data raises different challenges, including user queries understanding, multiple knowledge base support, and multilingual aspect. To address these challenges, we first design and develop an architecture to provide an interactive user interface. Secondly, we propose a machine learning approach based on intent classification and natural language understanding to understand user intents and generate SPARQL queries. We especially process a new social network dataset (i.e., myPersonality) and add it to the existing knowledge bases to extend the chatbot capabilities by understanding analytical queries. The system can be extended with a new domain on-demand, flexible, multiple knowledge base, multilingual, and allows intuitive creation and execution of different tasks for an extensive range of topics. Furthermore, evaluation and application cases in the chatbot are provided to show how it facilitates interactive semantic data towards different real application scenarios and showcase the proposed approach for a knowledge graph and data-driven chatbot.
DM-MCDA: Aweb-based platform for data mining and multiple criteria decision analysis: A case study on road accident.
Journal SoftwareX (2019) 100323, https://doi.org/10.1016/j.softx.2019.100323. 2019
Abstract
Today’s ultra-connected world is generating a huge amount of data stored in databases and cloud environment especially in the era of transportation. These databases need to be processed and analyzed to extract useful information and present it as a valid element for transportation managers for further use, such as road safety, shipping delays, and shipping optimization. The potential of data mining algorithms is largely untapped, this paper shows large-scale techniques such as associations rule analysis, multiple criteria analysis, and time series to improve road safety by identifying hot-spots in advance and giving chance to drivers to avoid the dangers. Indeed, we proposed a framework DM-MCDA based on association rules mining as a preliminary task to extract relationships between variables related to a road accident, and then integrate multiple criteria analysis to help decision-makers to make their choice of the most relevant rules. The developed system is flexible and allows intuitive creation and execution of different algorithms for an extensive range of road traffic topics. DM-MCDA can be expanded with new topics on demand, rendering knowledge extraction more robust and provide meaningful information that could help in developing suitable policies for decision-makers.
Stable routing protocol based on fuzzy logic system in vehicular ad hoc networks.
Journal Int J Commun Syst., 2018
Abstract
Vehicular ad‐hoc networks have several roles in alert messages dissemination between vehicles in danger, the most important role is to provide helpful information for drivers (eg, road traffic state). But, some performance improvements are frequently needed in terms of routing. Hence, several clustering approaches have been proposed to optimize the network services. These approaches are based on increasing data delivery, reducing data congestion, and dividing the traffic into clusters. However, a stable clustering algorithm is always required in order to ensure the data dissemination in a dense, mobile, or a large‐scale environment. Therefore, in this paper, we have proposed a stable routing protocol based on the fuzzy logic system, which can deliver alert messages with minimum delay and improve the stability of clusters structure by generating only a small number of clusters in the network. In this work, the fuzzy logic system has been used to create the clusters and select a cluster head for each cluster. We have used the network simulator (NS2) to generate the results. As a result, we could reduce the cluster head changes and increase the cluster member lifetime compared with recent approaches.
An improved approach for association rules mining using multi-criteria decision support system: A case study in road safety.
Journal European Transport Research Review, 2017.
Abstract
Road accidents have come to be considered a major public health problem worldwide. The aim of many studies is therefore to identify the main factors contributing to the severity of crashes. This paper examines a large-scale data mining technique known as association rule mining, which can predict future accidents in advance and allow drivers to avoid the dangers. However, this technique produces a very large number of decision rules, preventing decision makers from making their own selection of the most relevant rules. In this context, the integration of a multi-criteria decision analysis approach would be particularly useful for decision makers affected by the redundancy of the extracted rules. An analysis of road accidents in the province of Marrakech (Morocco) between 2004 and 2014 shows that the proposed approach serves this purpose; it may provide meaningful information that could help in developing suitable prevention policies to improve road safety.
Application of big data analysis with decision Tree for road accident.
Journal Indian Journal of Science and Technology, Vol. 10, No. 29, pp. 1-10, DOI: 10.17485/ijst/2017/v10i29/117325, 2017
Abstract
In transportation field, a huge amount of data collected by IoTsystems, remote sensing and other data collection tools brings new challenges, the size of this data becomes extremely big and more complex for traditional techniques of data mining. To deal with this challenge, Apache Spark stand as a powerful large scale distributed computing platform that can be used successfully for machine learning against very large databases. This work employed large-scale machine learning techniques especially Decision Tree with Apache Spark framework for big data analysis to build a model that can predict the factors lead to road accidents based on several input variables related to traffic accidents. Based on this, the predicting model first preprocesses the big accident data and analyze it to create data for a learning system. Empirical results show that the proposed model could provide new information that can assist the decision makers to analyze and improve road safety.
Intelligent transport system for road safety based data mining approach.
Journal International Journal of Control and Automation, Vol.10, No.8, pp.13-22, 2017
Abstract
Recently, road accidents considered a major public health problem worldwide, the aim of many studies is to identify the main factors that contribute to crash severity. To identify those factors this paper shows a large scale intelligent techniques, such as intelligent agents that can detect drivers’ cognitive state and analyze the data in a central system, the intelligent agents use data mining techniques, especially association rules mining to identify future accident in advance and giving chance to drivers to avoid the dangers. However, the association rule technique produces a huge amount of decision rules, which does not allow the decision makers to make their own selection of the most relevant rules. In this context, we believe that the visualization techniques would be particularly useful for decision makers who are suffering from the redundancy and quantity of extracted rules. An analysis of accidents on highways in the province of Marrakech (Morocco) between 2004 and 2014 showed that the proposed approach serves our purpose and may provide meaningful information that can help to develop suitable prevention policies to improve road safety
Mining and prioritization of association rules for Big Data: Multi-criteria decision analysis approach.
Journal Journal of Big Data, 2017
Abstract
Data mining techniques and extracting patterns from large datasets play a vital role in knowledge discovery. Most of the decision makers encounter a large number of decision rules resulted from association rules mining. Moreover, the volume of datasets brings a new challenge to extract patterns such as the cost of computing and inefficiency to achieve the relevant rules. To overcome these challenges, this paper aims to build a learning model based on FP-growth and Apache Spark framework to process and to extract relevant association rules. We also integrate the multi-criteria decision analysis to prioritize the extracted rules by taking into account the decision makers subjective judgment. We believe that this approach would be a useful model to follow, particularly for decision makers who are suffering from conflicts between extracted rules, and difficulties of building only the most interesting rules. Experimental results on road accidents analysis show that the proposed approach can be efficiently achieved more association rules with a higher accuracy rate and improve the response time of the proposed algorithm. The results make clear that the proposed approach performs well and can provide useful information that could help the decision makers to improve road safety.
A Choice of Symmetric Cryptographic Algorithms based on Multi-Criteria Analysis Approach for Securing Smart Grid.
Journal Indian Journal of Science and Technology, 2017
Abstract
In order to optimize production, consumption and distribution of energy, the different devices of a Smart Grid (SG) exchange daily increasing flows of information. Moreover, SG produces much more data stream than the traditional network. In addition to the large volume, the data of the SG are characterized by their diversity. However, securing these data flows is essential. Indeed, a single failure or attack could compromise the safety of the whole electrical network, the malfunction of which could have serious repercussions. Therefore, cryptography as a solution is necessary for SG to become realizable and secure. Being able to classify and to make a good choice of symmetric cryptographic algorithms for security of SG, we proposed to use an approach based on multi-criteria analysis.
Multi-criteria decisional approach for extracting relevant association rules.
Journal Int. J. of Computational Science and Engineering, 2017
Abstract
Association rule mining plays a vital role in knowledge discovery in databases. The difficult task is mining useful and non-redundant rules, in fact in most cases, the real datasets lead to a huge number of rules, which does not allow users to make their own selection of the most relevant. Several techniques are proposed such as rule clustering, informative cover method, quality measurements, etc. Another way to selecting relevant association rules, we believe it is necessary to integrate a decisional approach within the knowledge discovery process; to solve the problem, we propose an approach to discover a category of relevant association rules based on multi-criteria analysis MCA by using association rules as actions and quality measurements as criteria. Finally, we conclude our work by an empirical study to illustrate the performance of our proposed approach.
Multi-Agent Based Modeling for Extracting Relevant Association Rules Using a Multi-Criteria Analysis Approach.
Journal Vietnam Journal of Computer Science, Vol.3, N.4, pp 235–245, 2016
Abstract
Recently, association rule mining plays a vital role in knowledge discovery in database. In fact, in most cases, the real datasets lead to a very large number of rules, which do not allow users to make their own selection of the most relevant. The difficult task is mining useful and non-redundant rules. Several approaches have been proposed, such as rule clustering, informative cover method and quality measurements. Another way to selecting relevant association rules, we believe that it is necessary to integrate a decisional approach within the knowledge discovery process. Therefore, in this paper, we propose an approach to discover a category of relevant association rules based on multi-criteria analysis. In other side, the general process of association rules extraction becomes more and more complex, to solve such problem, we also proposed a multi-agent system for modeling the different process of our proposed approach. Therefore, we conclude our work by an empirical study applied to a set of banking data to illustrate the performance of our approach.
A Web-based Platform for Mining and Ranking Association Rules.
Conference ECIR The 42st European Conference on Information Retrieval, April 14-17, 2020, Lisbon, Spain.
Abstract
In this demo, we introduce an interactive system, which effectively applies multiple criteria analysis to rank association rules. We first use association rules techniques to explore the correlations between variables in given data (i.e., database and linked data (LD)), and secondly apply multiple criteria analysis (MCA) to select the most relevant rules according to user preferences. The developed system is flexible and allows intuitive creation and execution of different algorithms for an extensive range of advanced data analysis topics. Furthermore, we demonstrate a case study of association rule mining and ranking on road accident data.
Graph-based Interactive Data Federation System for Heterogeneous Data Retrieval and Analytics.
Conference The 30th TheWebConf'19 (formerly WWW), May 13-17, 2019. San Francisco, Californie, U.S.
Abstract
Given the increasing number of heterogeneous data stored in relational databases, file systems or cloud environment, it needs to be easily accessed and semantically connected for further data analytic. The potential of data federation is largely untapped, this paper presents an interactive data federation system (https://vimeo.com/319473546) by applying large-scale techniques including heterogeneous data federation, natural language processing, association rules and semantic web to perform data retrieval and analytics on social network data. The system first creates a Virtual Database (VDB) to virtually integrate data from multiple data sources. Next, a RDF generator is built to unify data, together with SPARQL queries, to support semantic data search over the processed text data by natural language processing (NLP). Association rule analysis is used to discover the patterns and recognize the most important co-occurrences of variables from multiple data sources. The system demonstrates how it facilitates interactive data analytic towards different application scenarios (e.g., sentiment analysis, privacy-concern analysis, community detection).
Improving RDF data through semantic association rules mining.
Conference The 31st annual workshop of the Swedish Artificial Intelligence Society (SAIS 2019), June 18-19, Umeå, Sweden.
Abstract
The ultra-connected world has been generating massive volumes of heterogeneous data stored in different data sources. And these data sources need to be normalized and interconnected to create a linked open data (LOD) that can be used to analyze, extract useful semantic knowledge, and present it as a valid element for decision making. Semantic web techniques (e.g., RDF, SPARQL) have been widely used for knowledge discovery. However, the primary issue of the semantic web is insufficiently integrated approaches, incompleteness, and incorrectness. To tackle this issue, we applied Natural Language Processing (NLP) and data mining to propose an approach for extracting semantic association rules from text stored in RDF data. The proposed approach is applied to mypersonality data and allows users to process status update, extract entities (NER), and then generate semantics transactions for traditional association rules algorithms.
Privacy-concerns in the Era of AI and Social Network Sites.
Conference The 31st annual workshop of the Swedish Artificial Intelligence Society (SAIS 2019), June 18-19, Umeå, Sweden.
Abstract
Social Network Sites (SNS) (e.g., Facebook, Youtube), have been playing a great role in our lives. On one hand, they help to connect people in the way that would otherwise never possible before. Many recent breakthroughs in AI such as facial recognition were achieved thanks to the amount of available data in the Internet via SNS. However, on the other hand, they can have major impacts on people life, good or bad. Due to privacy concerns, many people have tried to avoid SNS [2] to protect their privacy. Similar to the security issue of the Internet protocol, Machine Learning - which is the core of AI, was not designed with privacy in mind. For instance, Support Vector Machines (SVMs) try to solve a quadratic optimization problem by deciding which instances of training dataset are support vectors. This means that the data of people involved in the training process will be also published within the SVM models. Recently, Fredrikson et al. [1] used hill-climbing algorithm on the output probabilities of a computer-vision classifier to reveal individual faces from the training data. Because of all these issues, privacy guarantees must apply to the worst-case outliers and thus will also destroy data utilities. From all above reasons, in my PhD project, we study on (1) how to protect privacy when learning predictive models and how to have a good trade-off between data utilities and privacy, to avoid privacy breaches such as Cambridge Analytical in the future.
Privacy-aware Data Federation Infrastructure.
Conference The 31st annual workshop of the Swedish Artificial Intelligence Society (SAIS 2019), June 18-19, Umeå, Sweden.
Abstract
Social Network Sites (SNS) (e.g., Facebook, Youtube), have been playing a great role in our lives. On one hand, they help to connect people in the way that would otherwise never possible before. Many recent breakthroughs in AI such as facial recognition were achieved thanks to the amount of available data in the Internet via SNS. However, on the other hand, they can have major impacts on people life, good or bad. Due to privacy concerns, many people have tried to avoid SNS [2] to protect their privacy. Similar to the security issue of the Internet protocol, Machine Learning - which is the core of AI, was not designed with privacy in mind. For instance, Support Vector Machines (SVMs) try to solve a quadratic optimization problem by deciding which instances of training dataset are support vectors. This means that the data of people involved in the training process will be also published within the SVM models. Recently, Fredrikson et al. [1] used hill-climbing algorithm on the output probabilities of a computer-vision classifier to reveal individual faces from the training data. Because of all these issues, privacy guarantees must apply to the worst-case outliers and thus will also destroy data utilities. From all above reasons, in my PhD project, we study on (1) how to protect privacy when learning predictive models and how to have a good trade-off between data utilities and privacy, to avoid privacy breaches such as Cambridge Analytical in the future.
Application of big data analysis with decision Tree for road accident.
Conference 3rd International Conference on Green Computing and Engineering Technologies, August 8-10. Killaloe, County Clare, Ireland.
Abstract
Objectives: In transportation field, a huge amount of data collected by IoT systems, remote sensing and other data collection tools brings new challenges, the size of this data becomes extremely big and more complex for traditional techniques of data mining. To deal with this challenge, Apache Spark stand as a powerful large scale distributed computing platform that can be used successfully for machine learning against very large databases. This work employed large-scale machine learning techniques especially Decision Tree with Apache Spark framework for big data analysis to build a model that can predict the factors lead to road accidents based on several input variables related to traffic accidents. Based on this, the predicting model first preprocesses the big accident data and analyze it to create data for a learning system. Empirical results show that the proposed model could provide new information that can assist the decision makers to analyze and improve road safety
Intelligent transport system for road safety based data mining approach.
Conference 3rd International Conference on Green Computing and Engineering Technologies, August 8-10. Killaloe, County Clare, Ireland.
Abstract
Recently, road accidents considered a major public health problem worldwide, the aim of many studies is to identify the main factors that contribute to crash severity. To identify those factors this paper shows a large scale intelligent techniques, such as intelligent agents that can detect drivers’ cognitive state and analyze the data in a central system, the intelligent agents use data mining techniques, especially association rules mining to identify future accident in advance and giving chance to drivers to avoid the dangers. However, the association rule technique produces a huge amount of decision rules, which does not allow the decision makers to make their own selection of the most relevant rules. In this context, we believe that the visualization techniques would be particularly useful for decision makers who are suffering from the redundancy and quantity of extracted rules. An analysis of accidents on highways in the province of Marrakech (Morocco) between 2004 and 2014 showed that the proposed approach serves our purpose and may provide meaningful information that can help to develop suitable prevention policies to improve road safety.
An Approach Based on MapReduce and Decision Tree to Improve Road Safety in Morocco.
Conference International symposium on data engineering and information systems (DEIS'2017), Mai 19-05, 2017, Marrakech, Morocco.
Abstract
TBA
Mining Traffic Accident from Big Data: The case study of Morocco
Conference The First International Conference of High Innovation in Computer Science (ICHICS'2016), June 01-03, 2016, Kenitra, Morocco.
Abstract
TBA
An approach based on association rules mining to improve road safety in Morocco.
Conference International Conference on Information Technology for Organizations Development (IT4OD), Fez, 2016, pp. 1-6.doi: 10.1109/IT4OD.2016.7479311
Abstract
Traffic accident has a great impact on the socio-economic development of a society. this work employed large scale data mining method especially association rules and multi criteria analysis approach to discover new knowledge from historical data about traffic accidents in one of morocco busiest roads in order to assist police decision makers in the formulation of new policies and traffic rules on our highways management. The study focused on resulting from an accident using real data obtained from the Ministry of Equipment and Transport of morocco, Empirical results show that the developed models could provide new informations that can assist the authority to improve road safety.
An approach based on association rules mining to improve road safety in Morocco.
Conference International Conference on Information Technology for Organizations Development (IT4OD), Fez, 2016, pp. 1-6.doi: 10.1109/IT4OD.2016.7479311
Abstract
Traffic accident has a great impact on the socio-economic development of a society. this work employed large scale data mining method especially association rules and multi criteria analysis approach to discover new knowledge from historical data about traffic accidents in one of morocco busiest roads in order to assist police decision makers in the formulation of new policies and traffic rules on our highways management. The study focused on resulting from an accident using real data obtained from the Ministry of Equipment and Transport of morocco, Empirical results show that the developed models could provide new informations that can assist the authority to improve road safety.
Association Rules Mining based On Electre Tri Method.
Conference The 4th International Conference on Software Engineering and New Technologies (ICSENT'2015), December, 20-24, 2015, Istanbul, Turkey.
Abstract
TBA
A choice of relevant association rules based on multi-criteria analysis approach.
Conference 5th International Conference on Information and Communication Technology and Accessibility (ICTA), Marrakech, Morocco, 2015, pp. 1-6. doi: 10.1109/ICTA.2015.7426886
Abstract
The usefulness and relevance of association rules extracted by the generation algorithms are a critical problem. In fact, in most cases, the real datasets lead to a very large number of association rules, which does not allow users to make their own selection of the most relevant. The searching of the best from the vast array of extracted rules require the identification and use of good measures or techniques of choice. Partial panoramas of these are presented in numerous publications. In this context, we propose a new approach to selecting relevant categories of association rules based on multi criteria analysis using association rules as actions and measures as criteria.
Classification of routing Protocol in WSNs.
Conference The 4th International Conference on Software Engineering and New Technologies (ICSENT-2015), December, 20-24, 2015, Istanbul, Turkey.
Abstract
TBA
A comparative study of routing protocols in WSN.
Conference 5th International Conference on Information and Communication Technology and Accessibility (ICTA), Marrakech, Morocco, 2015, pp. 1-6.
Abstract
Wireless sensor network (WSN) is a technology that comes in response to the essential need of observing and controlling critical phenomenon. Nevertheless, this advanced technology is still limited due to many problems such as routing protocols, energy consumption, security and data aggregation. Being able to classify routing protocols, we proposed to use a new approach based on multiple criteria analysis in which we use protocols as actions and directed by decision maker's preferences.
Comparative survey of association rule mining algorithms based on multiple-criteria decision analysis approach.
Conference 3rd International Conference on Control, Engineering and Information Technology (CEIT), Tlemcen, Algeria 2015, pp. 1-6. doi: 10.1109/CEIT.2015.7233078
Abstract
Mining association rules is a leading task, which attracted the attention of researchers, it is one of the technical potential of data mining that allows discovered correlations and association between voluminous datasets. It generally spend two important steps, in the first is the extraction of frequent items, and extracting association rules from this frequent items for the second step. This extraction is a difficult task, costly in terms of response time and memory space as the number of frequent items is exponential to the number of items in database. Many algorithms have been designed to answer these problems. Nevertheless, the high number of algorithms is itself an obstacle to the ability of choice of an expert. In this context we propose an approach to make a good choice of extraction algorithm based on multi-criteria analysis.
Contact & Meet Me
I would be delighted to engage in a conversation with you, whether it's to offer my assistance in your research endeavors or provide support with your company's business strategies. I look forward to the opportunity to meet with you and explore potential collaborations on various projects. If you require additional information about my past work and experiences, please don't hesitate to reach out. I am more than willing to discuss these topics further at your convenience
- Phone: +46 0500-448333
- addi.ait-mlouk@his.se
- aitmlouk@gmail.com
- #aitmlouk
- addi-ait-mlouk
At My Office
I am readily available to assist you with your research inquiries and provide detailed insights into my previous work. My office is conveniently situated at the University of Skövde, where you can reach me during regular office hours, which are typically from 10:00 to 16:00 on most days. If you wish to discuss any specific topics or require additional information, please feel free to contact me via email or phone to schedule an appointment that suits your convenience.
His profileGallery →
No news!
Data Science, Machine Learning, Optimization and AI
- International Conference on Principles of Knowledge Representation and Reasoning (KR)
- Australasian Joint Conference on Artificial Intelligence (AI)
- International Conference on Machine Learning, Optimization, and Data Science (LOD)
- International Joint Conference on Neural Networks (IJCNN), IEEE
- International Symposium Advances in Artificial Intelligence and Applications (AAIA), IEEE
- SGAI International Conference on Artificial Intelligence, British Computer Society, Springer
- International Symposium on Applied Machine Intelligence and Informatics (SAMI), IEEE
- European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
- Neural Information Processing Systems (NeuroIPS)
- ACM International Conference on Information and Knowledge Management (CIKM)
- Conference on Uncertainty in Artificial Intelligence (UAI), Springer
- European Conference on Artificial Intelligence (ECAI), Springer
- Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD), Springer
- International Conference on Artificial Neural Networks (ICANN), Springer
- Pacific Rim International Conference on Artificial Intelligence (PRICAI), Springer
- International Conference on Neural Information Processing (ICONIP), Springer
- International Conference on Tools with Artificial Intelligence (ICTAI), IEEE
- International Joint Conference on Artificial Intelligence (IJCAI)
- International Conference on Autonomous Agents and Multi-agents Systems (AAMAS)
- AAAI Conference on Artificial Intelligence (AAAI)
- European Conference on Machine Learning (ECML)
- Asian Conference on Machine Learning (ACML)
- International Conference on Machine Learning (ICML)
- International Conference on Artificial Intelligence and Statistics (AISTATS)
- ACM Workshop on Artificial Intelligence and Security (ACM AISec)
- IEEE International Conference on Data Mining (IEEE ICDM)
- IEEE International Conference on Machine Learning and Applications (IEEE ICMLA)
Distributed Computing, Edge, Fog and Cloud Computing
- IEEE International Symposium on the Modeling, Analysis, and Simulation of Computer and Telecommunication Systems (IEEE MASCOTS)
- European Conference on Computer Systems (EuroSys)
- IEEE International Conference on Communications (IEEE ICC)
- IEEE International Conference on Data Science in Cyberspace (IEEE DSC)
- International Symposium on Research in Attacks, Intrusions and Defenses (RAID)
- IEEE Conference on Privacy, Security and Trust (IEEE PST)
- IEEE Symposium on Security and Privacy (IEEE S&P)
- IEEE Conference on Communications and Network Security (IEEE CNS)
- Network and Distributed System Security Symposium (NDSS)
- ACM Conference on Data and Application Security and Privacy (ACM CODAPSY)
- International Conference on Emerging Networking Experiments and Technologies (CoNEXT)
- International Conference on Information Networking (ICOIN)
- IEEE Global Communications Conference (IEEE GLOBECOM)
- IEEE International Conference on Computer Communications (IEEE INFOCOM)
- European Workshop on Systems Security (EuroSec)
- ACM Conference on Computer and Communications Security (ACM CCS)
- International Conference on Availability, Reliability and Security (ARES)
- IEEE European Symposium on Security and Privacy (IEEE EuroS&P)
- IEEE Conference on Dependable and Secure Computing (IEEE DSC)
- ACM ASIA Conference on Computer and Communications Security (ACM ASIACCS)
- IEEE International Conference on Trust, Security and Privacy in Computing and Communications (IEEE TrustCom)
- Australasian Conference on Information Security and Privacy (ACISP)