Pub Date : 2020-06-05DOI: 10.1109/CiSt49399.2021.9357248
Hanane Bouhmoud, D. Loudyi
Considering the significant operational and economic potential of Building Information Modeling (BIM), many countries have already set fundamental strategies toward its implementation including developing common BIM standards and guidelines leading to enhancing the productivity of Architecture, Engineering and Construction (AEC) industry and driving noticeable optimizations. However, despite the AEC industry development worldwide, several recent studies revealed that the level of BIM adoption in Africa is the lowest comparing to other continents. This paper aims to identify, classify and prioritize the barriers hindering BIM adoption and implementation in African countries through a literature review of existing challenges in African countries compared to countries in other continents. As result, this study reveals that Africa is facing further challenges that confine BIM technology to early stages of adoption.
{"title":"Building Information Modeling (BIM) barriers in Africa versus global challenges","authors":"Hanane Bouhmoud, D. Loudyi","doi":"10.1109/CiSt49399.2021.9357248","DOIUrl":"https://doi.org/10.1109/CiSt49399.2021.9357248","url":null,"abstract":"Considering the significant operational and economic potential of Building Information Modeling (BIM), many countries have already set fundamental strategies toward its implementation including developing common BIM standards and guidelines leading to enhancing the productivity of Architecture, Engineering and Construction (AEC) industry and driving noticeable optimizations. However, despite the AEC industry development worldwide, several recent studies revealed that the level of BIM adoption in Africa is the lowest comparing to other continents. This paper aims to identify, classify and prioritize the barriers hindering BIM adoption and implementation in African countries through a literature review of existing challenges in African countries compared to countries in other continents. As result, this study reveals that Africa is facing further challenges that confine BIM technology to early stages of adoption.","PeriodicalId":253233,"journal":{"name":"2020 6th IEEE Congress on Information Science and Technology (CiSt)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120977734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/CiSt49399.2021.9357176
Anne-Carole Honfoga, M. Dossou, Péniel Dassi, V. Moeyaert
The Digital Video Broadcasting-Terrestrial, second generation (DVB-T2) system has been the object of many research during the last decade and is now mature. This paper focuses on the identification of the maximum obtainable performance improvement of DVB-T2 CP-OFDM with NUCs in typical TU6 fading environment. It first concentrates on the ultimate improvement achievable using joint UFMC and NUCs in standardized DVB-T2 system. In these conditions, a 0.5 dB SNR improvement (for $text{BER}=3.10^{-3}$) is reported in TU6 using CP-OFDM NUC 32K 256-QAM and $text{CR}=1/2$ and 3/5 in place of sole CP-OFDM. Also, using in conjunction both studied technologies, namely UFMC NUC $32K$ 256-QAM $text{CR}=1/2$ and 3/5, an SNR improvement of 1.2 dB (for $text{BER}=3.10^{-3}$) is achievable which provides a good SNR margin e.g. to increase the emitter's coverage.
{"title":"Joint use of 5G waveform UFMC and Non Uniform Constellations in DVB-T2","authors":"Anne-Carole Honfoga, M. Dossou, Péniel Dassi, V. Moeyaert","doi":"10.1109/CiSt49399.2021.9357176","DOIUrl":"https://doi.org/10.1109/CiSt49399.2021.9357176","url":null,"abstract":"The Digital Video Broadcasting-Terrestrial, second generation (DVB-T2) system has been the object of many research during the last decade and is now mature. This paper focuses on the identification of the maximum obtainable performance improvement of DVB-T2 CP-OFDM with NUCs in typical TU6 fading environment. It first concentrates on the ultimate improvement achievable using joint UFMC and NUCs in standardized DVB-T2 system. In these conditions, a 0.5 dB SNR improvement (for $text{BER}=3.10^{-3}$) is reported in TU6 using CP-OFDM NUC 32K 256-QAM and $text{CR}=1/2$ and 3/5 in place of sole CP-OFDM. Also, using in conjunction both studied technologies, namely UFMC NUC $32K$ 256-QAM $text{CR}=1/2$ and 3/5, an SNR improvement of 1.2 dB (for $text{BER}=3.10^{-3}$) is achievable which provides a good SNR margin e.g. to increase the emitter's coverage.","PeriodicalId":253233,"journal":{"name":"2020 6th IEEE Congress on Information Science and Technology (CiSt)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125022709","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/CiSt49399.2021.9357258
Luigi Bambaci
Manually encoding variant readings is a difficult and time-consuming task. Markup languages ensure data exchange and reusability but are very difficult to handle especially in the case of texts characterized by a rich textual tradition and editions with extensive critical apparatus. Scholars engaged in digitizing printed critical editions find themselves dealing with different levels of problems, including the revision of OCR outputs and the conversion from plain text to a coherent XML encoding. In this article we illustrate how it is possible to exploit the structured language of critical apparatus and the conventions of the domain of textual philology as means to automate processing and encoding. Finally we discuss the advantages deriving from the adoption of a parsing system over a manual encoding, which go from data compression to the possibility of automatically detecting inconsistencies in the printed source, of correcting errors originated after OCR processing, and of better controlling the generation of semantic errors during conversion into XML code. Our case study concerns the digitization of a collation of Hebrew manuscripts and printed editions realized by the English scholar Benjamin Kennicott in the second half of the XVIII century.
{"title":"Digitizing an Eighteenth Century Collation of Hebrew Manuscripts: A Rule-Based Parsing System for Automatically Encoding Critical Apparatus","authors":"Luigi Bambaci","doi":"10.1109/CiSt49399.2021.9357258","DOIUrl":"https://doi.org/10.1109/CiSt49399.2021.9357258","url":null,"abstract":"Manually encoding variant readings is a difficult and time-consuming task. Markup languages ensure data exchange and reusability but are very difficult to handle especially in the case of texts characterized by a rich textual tradition and editions with extensive critical apparatus. Scholars engaged in digitizing printed critical editions find themselves dealing with different levels of problems, including the revision of OCR outputs and the conversion from plain text to a coherent XML encoding. In this article we illustrate how it is possible to exploit the structured language of critical apparatus and the conventions of the domain of textual philology as means to automate processing and encoding. Finally we discuss the advantages deriving from the adoption of a parsing system over a manual encoding, which go from data compression to the possibility of automatically detecting inconsistencies in the printed source, of correcting errors originated after OCR processing, and of better controlling the generation of semantic errors during conversion into XML code. Our case study concerns the digitization of a collation of Hebrew manuscripts and printed editions realized by the English scholar Benjamin Kennicott in the second half of the XVIII century.","PeriodicalId":253233,"journal":{"name":"2020 6th IEEE Congress on Information Science and Technology (CiSt)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129185567","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/CiSt49399.2021.9357177
Salma Saidani, R. Haddad, R. Bouallègue
M-health technology is a key component in digital healthcare monitoring. The Pedology domain has become a thriving topic in the last years. Researches prove that there is a relation between feet analysis and longer healthy life. In this paper, we propose a design of a smart insole system equipped with four resistive sensors (FSR) to provide real-time control of patients without bothering them during their daily activities. The pressure sensors were placed at the bottom of the shoe to assess timing parameters and pressure distribution. Most of the body pressure is measured from the rearfoot and from the forefoot regions. Considering them we have placed two of our sensors in the forefoot region in the hallux part and the last one in the heel. The microcontroller collects the physiological parameters information from the sensors and sends them to a mobile application via Bluetooth interface to allow the user to monitor his gait cycle over time and receive in real-time notification. The data is saved on a database and restored to a web application from which the clinician controls the physical activity of the patients and intervenes in the best moment to help him, then all the data it will be saved on in a firebase. Our proposal proves good results compared to other related works.
{"title":"A prototype design of a smart shoe insole system for real-time monitoring of patients","authors":"Salma Saidani, R. Haddad, R. Bouallègue","doi":"10.1109/CiSt49399.2021.9357177","DOIUrl":"https://doi.org/10.1109/CiSt49399.2021.9357177","url":null,"abstract":"M-health technology is a key component in digital healthcare monitoring. The Pedology domain has become a thriving topic in the last years. Researches prove that there is a relation between feet analysis and longer healthy life. In this paper, we propose a design of a smart insole system equipped with four resistive sensors (FSR) to provide real-time control of patients without bothering them during their daily activities. The pressure sensors were placed at the bottom of the shoe to assess timing parameters and pressure distribution. Most of the body pressure is measured from the rearfoot and from the forefoot regions. Considering them we have placed two of our sensors in the forefoot region in the hallux part and the last one in the heel. The microcontroller collects the physiological parameters information from the sensors and sends them to a mobile application via Bluetooth interface to allow the user to monitor his gait cycle over time and receive in real-time notification. The data is saved on a database and restored to a web application from which the clinician controls the physical activity of the patients and intervenes in the best moment to help him, then all the data it will be saved on in a firebase. Our proposal proves good results compared to other related works.","PeriodicalId":253233,"journal":{"name":"2020 6th IEEE Congress on Information Science and Technology (CiSt)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130325608","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/CiSt49399.2021.9357175
Zahra Berradi, M. Lazaar, O. Mahboub, Hicham Omara
Artificial intelligence is a vast and promising domain that helps improving people's lives in different areas such as, medicine, education, telecommunication, finance, and economy. The financial market is an important aspect of the economics of any country, by having a clear idea about how it functions, it helps to improve the economy of the country radically and, therefore, the people's lives. In this paper, we suggest giving the latest research of deep learning techniques applied on the financial market field that can help investors to make an accurate decision. This paper gathered all the recent articles related to deep learning techniques applied on forecasting the financial market, which includes stock market, stock index, commodity forecasting and Forex. The main goal is to find the most models used recently to solve the prediction problem using RNN, their characteristic and novelty. We will give all aspects that involve the process of forecasting beginning with preprocessing, the input features, the deep learning techniques, and the evaluation metrics employed.
{"title":"A Comprehensive Review of Artificial Intelligence Techniques in Financial Market","authors":"Zahra Berradi, M. Lazaar, O. Mahboub, Hicham Omara","doi":"10.1109/CiSt49399.2021.9357175","DOIUrl":"https://doi.org/10.1109/CiSt49399.2021.9357175","url":null,"abstract":"Artificial intelligence is a vast and promising domain that helps improving people's lives in different areas such as, medicine, education, telecommunication, finance, and economy. The financial market is an important aspect of the economics of any country, by having a clear idea about how it functions, it helps to improve the economy of the country radically and, therefore, the people's lives. In this paper, we suggest giving the latest research of deep learning techniques applied on the financial market field that can help investors to make an accurate decision. This paper gathered all the recent articles related to deep learning techniques applied on forecasting the financial market, which includes stock market, stock index, commodity forecasting and Forex. The main goal is to find the most models used recently to solve the prediction problem using RNN, their characteristic and novelty. We will give all aspects that involve the process of forecasting beginning with preprocessing, the input features, the deep learning techniques, and the evaluation metrics employed.","PeriodicalId":253233,"journal":{"name":"2020 6th IEEE Congress on Information Science and Technology (CiSt)","volume":"2 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130272939","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/CiSt49399.2021.9357289
E. Kenane, M. Garah, F. Benmeddour
In the present paper, a dual band four-way Wilkinson power divider (WPD) using a genetic algorithm (GA) optimization is designed and simulated using ADS Momentum simulator. The proposed power divider is developed to support dual bands of WLAN applications. The present power divider is formed by many transmission lines sections. The transmission lines are printed on FR4 (Flame Resistant 4) substrate of dielectric constant $varepsilon_{mathrm{r}}$ of3.4 with loss tangent of 0.0023. The proposed circuit exhibits good characteristics at WLAN frequencies (2.45GHz & 5.8 GHZ) in terms of return loss and isolation.
{"title":"A Dual Band Four Ports WILKINSON Power Divider Design","authors":"E. Kenane, M. Garah, F. Benmeddour","doi":"10.1109/CiSt49399.2021.9357289","DOIUrl":"https://doi.org/10.1109/CiSt49399.2021.9357289","url":null,"abstract":"In the present paper, a dual band four-way Wilkinson power divider (WPD) using a genetic algorithm (GA) optimization is designed and simulated using ADS Momentum simulator. The proposed power divider is developed to support dual bands of WLAN applications. The present power divider is formed by many transmission lines sections. The transmission lines are printed on FR4 (Flame Resistant 4) substrate of dielectric constant $varepsilon_{mathrm{r}}$ of3.4 with loss tangent of 0.0023. The proposed circuit exhibits good characteristics at WLAN frequencies (2.45GHz & 5.8 GHZ) in terms of return loss and isolation.","PeriodicalId":253233,"journal":{"name":"2020 6th IEEE Congress on Information Science and Technology (CiSt)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129501660","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/CiSt49399.2021.9357181
Saad Chakkor, Mostafa Baghouri, Zineb Cheker, A. Oualkadi, J. E. Hangouche, Jawhar Laamech
This is a proposal for an automated detection and remote monitoring system made up of a centralized network of communicating portable electronic devices based on biomedical sensors operating in the IoT context in synergy with wireless sensor network technologies, telemedicine and artificial intelligence. This network will be deployed to monitor a population settling in a target area (cities, region, country, etc.). The goal of this system is the detection and early diagnosis of the disease in people infected with the COVID-19 virus, using a device (such as a bracelet or a chest strap). This device collects in real time all the necessary biomedical measurements of a person, including their location, freeing them from any hospitalization or use of complex and expensive equipment. These informations are then transmitted, via a wireless connection, to a regional or national control center which takes care of its storage in a specialized database. This center executes a decision-making algorithm using artificial intelligence and fuzzy inference engine to detect accurately each possible abnormal change in the supervised biomedical signs reflecting risk factor or indicating the appearance of symptoms characterizing COVID-19 disease. In the positive case, the control system triggers a warning alarm concerning this infected person and requests intervention of the competent authorities to take the necessary measures and actions. Computer simulations with Matlab software tool have been conducted to evaluate the performance of the proposed system. Study findings show that the designed device is suitable for application in COVID-19 patient monitoring.
{"title":"Intelligent Network for Proactive Detection of COVID-19 Disease","authors":"Saad Chakkor, Mostafa Baghouri, Zineb Cheker, A. Oualkadi, J. E. Hangouche, Jawhar Laamech","doi":"10.1109/CiSt49399.2021.9357181","DOIUrl":"https://doi.org/10.1109/CiSt49399.2021.9357181","url":null,"abstract":"This is a proposal for an automated detection and remote monitoring system made up of a centralized network of communicating portable electronic devices based on biomedical sensors operating in the IoT context in synergy with wireless sensor network technologies, telemedicine and artificial intelligence. This network will be deployed to monitor a population settling in a target area (cities, region, country, etc.). The goal of this system is the detection and early diagnosis of the disease in people infected with the COVID-19 virus, using a device (such as a bracelet or a chest strap). This device collects in real time all the necessary biomedical measurements of a person, including their location, freeing them from any hospitalization or use of complex and expensive equipment. These informations are then transmitted, via a wireless connection, to a regional or national control center which takes care of its storage in a specialized database. This center executes a decision-making algorithm using artificial intelligence and fuzzy inference engine to detect accurately each possible abnormal change in the supervised biomedical signs reflecting risk factor or indicating the appearance of symptoms characterizing COVID-19 disease. In the positive case, the control system triggers a warning alarm concerning this infected person and requests intervention of the competent authorities to take the necessary measures and actions. Computer simulations with Matlab software tool have been conducted to evaluate the performance of the proposed system. Study findings show that the designed device is suitable for application in COVID-19 patient monitoring.","PeriodicalId":253233,"journal":{"name":"2020 6th IEEE Congress on Information Science and Technology (CiSt)","volume":"252 12","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120862631","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/CiSt49399.2021.9357192
A. D. Grosso, Driss Fassi Fihri, M. E. Mohajir, Ouafae Nahli, A. Tonazzini
In this paper, we analyze and discuss the characteristics of a system for the effective digital preservation and fruition of historical manuscripts degraded by the process of lamination. As a case study, we will make reference to the “Poem in Rajaz on medicine”, written by Abubacer in the XII century, and conserved in the Al Quaraouiyine Library located in Fez, Morocco. The conceived system should have at least four main functionalities: image acquisition (i.e. digitization), image enhancement, text encoding, and linguistic analysis. Based on the evaluation of the manuscript damages, the acquisition set up should be designed in such a way to be able to avoid reflections as much as possible. Suitable digital image processing techniques should also be devised to correct the residual degradations and enhance the text for an easier legibility. Finally, semi-automatic transcription, scholarly encoding and linguistic analysis, to be performed on the virtually restored pages, should adapt existing tools to the specificity of the primary source writing system and language. The feasibility study for the realization of such a system is of general utility, in that it can provide guidelines for the digitization, the enhancement and the text encoding of the many laminated manuscripts conserved in other historical archives. On the other hand, from the cultural heritage point of view, the experimentation on the “Poem in Rajaz on medicine” could foster the systematic philological and ontological study of a unique piece of our documental heritage: the longest poem of medieval Islamic medical literature.
{"title":"Digital safeguard of laminated historical manuscripts: the treatise “Poem in Rajaz on medicine” as a case study","authors":"A. D. Grosso, Driss Fassi Fihri, M. E. Mohajir, Ouafae Nahli, A. Tonazzini","doi":"10.1109/CiSt49399.2021.9357192","DOIUrl":"https://doi.org/10.1109/CiSt49399.2021.9357192","url":null,"abstract":"In this paper, we analyze and discuss the characteristics of a system for the effective digital preservation and fruition of historical manuscripts degraded by the process of lamination. As a case study, we will make reference to the “Poem in Rajaz on medicine”, written by Abubacer in the XII century, and conserved in the Al Quaraouiyine Library located in Fez, Morocco. The conceived system should have at least four main functionalities: image acquisition (i.e. digitization), image enhancement, text encoding, and linguistic analysis. Based on the evaluation of the manuscript damages, the acquisition set up should be designed in such a way to be able to avoid reflections as much as possible. Suitable digital image processing techniques should also be devised to correct the residual degradations and enhance the text for an easier legibility. Finally, semi-automatic transcription, scholarly encoding and linguistic analysis, to be performed on the virtually restored pages, should adapt existing tools to the specificity of the primary source writing system and language. The feasibility study for the realization of such a system is of general utility, in that it can provide guidelines for the digitization, the enhancement and the text encoding of the many laminated manuscripts conserved in other historical archives. On the other hand, from the cultural heritage point of view, the experimentation on the “Poem in Rajaz on medicine” could foster the systematic philological and ontological study of a unique piece of our documental heritage: the longest poem of medieval Islamic medical literature.","PeriodicalId":253233,"journal":{"name":"2020 6th IEEE Congress on Information Science and Technology (CiSt)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126772028","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/CiSt49399.2021.9357249
Oussama Chabih, Sara Sbai, Hicham Behja, Mohammed Reda Chbihi Louhdi, E. Zemmouri, B. Trousse
One of the most common problems in the field of information processing is the definition of the optimal number of clusters when using unsupervised classification. The wrong choice of the number of clusters leads to insignificant and inconsistent results. In this paper, we present a simple automatic technique for estimating the optimal number of clusters when using a classification algorithm based on k-means. The proposed algorithm is based on detecting the immutable clusters during the different experiments and takes them as a reference in order to specify the best classification.
{"title":"New approach to determine the optimal number of clusters K in unsupervised classification","authors":"Oussama Chabih, Sara Sbai, Hicham Behja, Mohammed Reda Chbihi Louhdi, E. Zemmouri, B. Trousse","doi":"10.1109/CiSt49399.2021.9357249","DOIUrl":"https://doi.org/10.1109/CiSt49399.2021.9357249","url":null,"abstract":"One of the most common problems in the field of information processing is the definition of the optimal number of clusters when using unsupervised classification. The wrong choice of the number of clusters leads to insignificant and inconsistent results. In this paper, we present a simple automatic technique for estimating the optimal number of clusters when using a classification algorithm based on k-means. The proposed algorithm is based on detecting the immutable clusters during the different experiments and takes them as a reference in order to specify the best classification.","PeriodicalId":253233,"journal":{"name":"2020 6th IEEE Congress on Information Science and Technology (CiSt)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126776642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-05DOI: 10.1109/CiSt49399.2021.9357298
S. Elyassami, Yasir Hamid, T. Habuza
People lose their lives every day due to road traffic crashes. The problem is so humongous globally that the World Health Organization, in its Sustainable Development Agenda 2030, is inviting the coordinates efforts across nations towards it and aspiring to cut down the deaths and injuries to half. Taking a clue from that, the proposed work is undertaken to build machine learning-based models for analyzing the crash data, identifying the important risk factors, and predict the injury severity of drivers. The proposed work studied and analyzed several factors of road accidents to create an accurate and interpretable model that predicts the occurrence and severity of car accidents by investigating crash causal factors and crash severity factors. In the proposed work, we employed three machine learning algorithms to vis-à-vis Decision Tree, Random Forest, and Gradient Boosted tree on Statewide Vehicle Crashes Dataset provided by Maryland State Police. The gradient boosted-based model reported the highest prediction accuracy and provided the most influencing factors in the predictive model. The findings showed that disregarding traffic signals and stop signs, road design problems, poor visibility, and bad weather conditions are the most important variables in the predictive road traffic crash model. Using the identified risk factors is crucial in establishing actions that may reduce the risks related to those factors.
{"title":"Road Crashes Analysis and Prediction using Gradient Boosted and Random Forest Trees","authors":"S. Elyassami, Yasir Hamid, T. Habuza","doi":"10.1109/CiSt49399.2021.9357298","DOIUrl":"https://doi.org/10.1109/CiSt49399.2021.9357298","url":null,"abstract":"People lose their lives every day due to road traffic crashes. The problem is so humongous globally that the World Health Organization, in its Sustainable Development Agenda 2030, is inviting the coordinates efforts across nations towards it and aspiring to cut down the deaths and injuries to half. Taking a clue from that, the proposed work is undertaken to build machine learning-based models for analyzing the crash data, identifying the important risk factors, and predict the injury severity of drivers. The proposed work studied and analyzed several factors of road accidents to create an accurate and interpretable model that predicts the occurrence and severity of car accidents by investigating crash causal factors and crash severity factors. In the proposed work, we employed three machine learning algorithms to vis-à-vis Decision Tree, Random Forest, and Gradient Boosted tree on Statewide Vehicle Crashes Dataset provided by Maryland State Police. The gradient boosted-based model reported the highest prediction accuracy and provided the most influencing factors in the predictive model. The findings showed that disregarding traffic signals and stop signs, road design problems, poor visibility, and bad weather conditions are the most important variables in the predictive road traffic crash model. Using the identified risk factors is crucial in establishing actions that may reduce the risks related to those factors.","PeriodicalId":253233,"journal":{"name":"2020 6th IEEE Congress on Information Science and Technology (CiSt)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132751878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：