• Issue 1 (2020)

IoT and Big Data Analytics for Smart Buildings: A Survey

JUSPN, volume-13 , Issue 1 (2020), PP 27 - 34

Published: 23 Aug 2020

DOI: 10.5383/JUSPN.13.01.004

by A Daissaoui, A Boulmakoul, L Karim d , A Lbath from LIG/ MRIM, CNRS, University Grenoble Alpes, France EMSI, LPRI, Casablanca, Morocco LIM/IOS., FSTM, Hassan II University of Casablanca, Morocco ENSA Berrechid, Hassan 1st University, Morocco

Abstract: The processes of digital transformation have involved a variety of socio-technical activities, with the objective of increasing productivity, safety and quality of execution, sustainable development, collaborative working, and solutions for the sustainable smart buildings. The major technological trends have changed the building sector and revealing new understanding on how data generated by all these new technologies can be used to achieve the essence of smart building objective. Current smart building management systems incorporate a variety of IoT technologies, a structured data storage environment and analytics data tools. Their objectives are to observe the condition of specific areas and apply appropriate rules to preserve or improve comfort while saving energy. In this paper, we propose a review of related works to IoT, Big Data analytics in smart buildings. read more... read less...

Keywords: Smart buildings, IoT, Big data analytics, Reactif systems, Complex event processing

Student orientation using machine learning under MapReduce with Hadoop

JUSPN, volume-13 , Issue 1 (2020), PP 21 - 26

Published: 21 Aug 2020

DOI: 10.5383/JUSPN.13.01.003

by F.Ouatika, M.Erritali , F.Ouatik, M.Jourhmaned from Department of Computers Sciences, Sultan Moulay Slimane University,Beni Mellal, Morocco

Abstract: Academic orientation is a procedure which consists in helping students to succeed in their educational path and to know the ideal path. the choice, of course, allows the student to prepare for professional life and find the right path to the future. It is therefore essential to choose the sector well to avoid any pitfall and risk hazardous choices and also reorientation. There are establishments which organize orientation sessions with the students, but the number of students is very high, which pushed us to carry out an orientation system to make this service accessible by all the students and also to make the procedure very powerful orientation and in a lapse of time. Decision making related to orientation is very complicated by the fact that it is linked to several factors (prerequisite, the student's previous academic career, marks and the number of absences by subject) as the learner’s marks give us information about his abilities and the extent of his mastery of the specialization subjects. As for the student’s tendencies and desires, they appear through his keenness on these subjects, and he is not absent from them, And it is evidenced by the number of absences by subject of the learner.These data are useful, but they pose problems in storage and processing, so the solution is to use Big Data technology. In this article we used the notes and the number of student absences by subject. These data are stored in Hadoop Distributed File System (HDFS) and processed by MapReduce using Hadoop framework. We compared the classification accuracy and speed up of Neural Networks, Naive Bayes and k-nearest-neighbors classification algorithms to make the decision and we found that Naive Bayes is the most suitable for this procedure. read more... read less...

Keywords: Machine learning, Big Data, MapReduce, K-Nearest-Neighbors, Neural Networks and Naïve Bayes.

QoS-Aware Placement of Tasks on a Fog Cluster in an Edge Computing Environment

JUSPN, volume-13 , Issue 1 (2020), PP 11 - 19

Published: 14 Aug 2020

DOI: 10.5383/JUSPN.13.01.002

by Elarbi Badidi from Department of Computer Science and Software Engineering, College of Information Technology, UAE University, AL-AIN, PO Box. 15551, UAE

Abstract: The advances made in the sensing and communications technologies over the last few years have made the deployment of IoT solutions possible on a massive scale. The wide deployment of IoT sensors and devices has resulted in the development of smart services that were not possible before. These services typically rely on cloud services for processing IoT data streams, given that edge devices have limited computing and storage capabilities. However, time-sensitive IoT applications and services do not tolerate the high latency they can encounter when sending IoT data streams to the cloud. Fog computing-based solutions for these services are increasingly becoming attractive because of the low latency they can guarantee. With increasing deployments of fog nodes and fog clusters, we propose an architecture for the placement of IoT applications tasks on a cluster of fog nodes in the vicinity of the application’s data sources. The Fog Broker component can implement various scheduling policies to help IoT applications meet their quality-of-service (QoS) requirements. Our simulations show that it is possible to maintain low application latency and distribute the load between the fog nodes of the cluster by using simple scheduling strategies. read more... read less...

Keywords: Fog computing, cloud computing, quality-of-service, latency, scheduling

Usability Study of a comprehensive table tennis AR-based training system with the focus on players' strokes

JUSPN, volume-13 , Issue 1 (2020), PP 01 - 09

Published: 11 Aug 2020

DOI: 10.5383/JUSPN.13.01.001

by Ayman Nabil, Habiba Hegazy, Mohamed Abdelsalam, Moustafa Hussien, Seif Elmosalamy, Yomna M.I. Hassan, Ayman Atia from Faculty of Computer Science, Misr International University, Egypt, HCI-LAB, Faculty of Computers and Artificial Intelligence, Helwan University, Egypt, October University for Modern Sciences and Arts (MSA), Egypt

Abstract: Table tennis game is based on the speed of the player’s response to different attacks and defense strokes. A way to enhance the player’s performance and technique while training is to update the player with the mistakes in real-time. This paper presents a system that focuses on detecting the correct and wrong strokes within the following stroke types: Forehand drive, backhand drive, and forehand topspin. By the usage of Augmented Reality, the system helps the players to get their results and direction easily using AR-based mobile application when practicing real-time. A usability study has been made to measure the learning style of the players by letting the players train on different strokes with the system. Moreover, an experiment has been done to measure the efficiency of the application and compare different algorithms to overview their performance in identifying the strokes based on accuracy and time taken. read more... read less...

Keywords: Stroke Classification, Stroke Identification, Table Tennis, IR Depth Camera, Hand Gestures, Augmented Reality