Pub Date : 2018-12-01DOI: 10.1109/ICONIC.2018.8601086
G. Mosweunyane, N. Motlogelwa, G. Malema
Model curricula such as ACM/IEEE recommend that characteristics of students graduating in Computer Science should include project experience. In line with these recommendations, students in their fourth year of undergraduate study in the Computer Science department at the University of Botswana have to complete a major project as part of fulfilling the requirements for the award of a degree. The project gives students the opportunity to implement some of the theories and techniques learnt during the course of their programme of study. Students‘ choice of project topics therefore heavily depends on what the curriculum offers in terms of diversity of course content and techniques taught. This paper presents an analysis of 154 students’ projects in the Computer Science department at the University of Botswana for the period 2013–2016 with a particular focus on the topic trends and issues of diversity and relevance to both the local and international context. The topics’ list is categorized using the Agglomerative Hierarchical Clustering method. The analysis of titles and abstracts shows an inclination towards web-based systems and mobile applications that are relevant to the local context. Also, the titles and abstracts do not reveal core Computer Science areas such as algorithms, machine learning and security. Trending topics in Computing, which are also typically done as project topics, also seem not to feature. To reveal an effective categorization and assess comparability to international trends the rest of entire project document (body) needs to be analysed.
{"title":"A Survey of Undergraduate Project Topics in Computer Science at the University of Botswana","authors":"G. Mosweunyane, N. Motlogelwa, G. Malema","doi":"10.1109/ICONIC.2018.8601086","DOIUrl":"https://doi.org/10.1109/ICONIC.2018.8601086","url":null,"abstract":"Model curricula such as ACM/IEEE recommend that characteristics of students graduating in Computer Science should include project experience. In line with these recommendations, students in their fourth year of undergraduate study in the Computer Science department at the University of Botswana have to complete a major project as part of fulfilling the requirements for the award of a degree. The project gives students the opportunity to implement some of the theories and techniques learnt during the course of their programme of study. Students‘ choice of project topics therefore heavily depends on what the curriculum offers in terms of diversity of course content and techniques taught. This paper presents an analysis of 154 students’ projects in the Computer Science department at the University of Botswana for the period 2013–2016 with a particular focus on the topic trends and issues of diversity and relevance to both the local and international context. The topics’ list is categorized using the Agglomerative Hierarchical Clustering method. The analysis of titles and abstracts shows an inclination towards web-based systems and mobile applications that are relevant to the local context. Also, the titles and abstracts do not reveal core Computer Science areas such as algorithms, machine learning and security. Trending topics in Computing, which are also typically done as project topics, also seem not to feature. To reveal an effective categorization and assess comparability to international trends the rest of entire project document (body) needs to be analysed.","PeriodicalId":277315,"journal":{"name":"2018 International Conference on Intelligent and Innovative Computing Applications (ICONIC)","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125548302","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 : 2018-12-01DOI: 10.1109/ICONIC.2018.8601232
Bhekisisa Nyoni, M. Nleya, B. Mtunzi
This prototype system known as the Wearable Instantaneous Ball Speed Estimator (WIBASE) was designed to measure the bowling speed of a cricketer during training. When fast bowlers are training, coaches have to assess their ability to bowl consistently fast balls even when they are required to perform long bowling spells, hence the need for reliable, accessible and affordable equipment for measuring their bowling speed cannot be over emphasised. The WIBASE seeks to fill in this gap. It is made up of two hardware components; a computer and a wrist-worn electronic board that houses among other components, a 3-dimensional (3D) acceleration sensor. The system tracks the three-axis acceleration generated by the movement of the arm when delivering the ball and stores these values. The raw sensor data from three different sensors namely accelerometer, gyroscope and magnetometer is processed by a Digital Motion Processor (DMP) on the board in a process known as Sensor Fusion before it is sent via Bluetooth to the computer. The computer runs a Python script that receives the filtered acceleration which consists of both static acceleration and dynamic acceleration. The acceleration is numerically integrated over a minute period of time around the release point using the Trapezoidal method of integrating numerical data to derive the speed of the cricket bowler. The results obtained from the three sets of experiments that were conducted show that the WIBASE can track the 3D acceleration of the hand when bowling, derive the speed of the bowlers and display the speed on a computer while logging all the data into a file.
{"title":"A Training Utility for Estimating the Bowling Speed of a Cricketer Using Accelerometer Data","authors":"Bhekisisa Nyoni, M. Nleya, B. Mtunzi","doi":"10.1109/ICONIC.2018.8601232","DOIUrl":"https://doi.org/10.1109/ICONIC.2018.8601232","url":null,"abstract":"This prototype system known as the Wearable Instantaneous Ball Speed Estimator (WIBASE) was designed to measure the bowling speed of a cricketer during training. When fast bowlers are training, coaches have to assess their ability to bowl consistently fast balls even when they are required to perform long bowling spells, hence the need for reliable, accessible and affordable equipment for measuring their bowling speed cannot be over emphasised. The WIBASE seeks to fill in this gap. It is made up of two hardware components; a computer and a wrist-worn electronic board that houses among other components, a 3-dimensional (3D) acceleration sensor. The system tracks the three-axis acceleration generated by the movement of the arm when delivering the ball and stores these values. The raw sensor data from three different sensors namely accelerometer, gyroscope and magnetometer is processed by a Digital Motion Processor (DMP) on the board in a process known as Sensor Fusion before it is sent via Bluetooth to the computer. The computer runs a Python script that receives the filtered acceleration which consists of both static acceleration and dynamic acceleration. The acceleration is numerically integrated over a minute period of time around the release point using the Trapezoidal method of integrating numerical data to derive the speed of the cricket bowler. The results obtained from the three sets of experiments that were conducted show that the WIBASE can track the 3D acceleration of the hand when bowling, derive the speed of the bowlers and display the speed on a computer while logging all the data into a file.","PeriodicalId":277315,"journal":{"name":"2018 International Conference on Intelligent and Innovative Computing Applications (ICONIC)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126377610","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 : 2018-12-01DOI: 10.1109/ICONIC.2018.8601255
Olawuyi O. Fatoki, S. Ojo, P. Owolawi, T. Mapayi
Glaucoma, one of the major causes of blindness, has been identified as a disease that causes the degeneration of the optic disc. An highly accurate automated detection of the optic disc (OD) has however been noted to be of great importance for the efficient diagnosis of the disease. This paper presents a study on an unsupervised approach usage of Haralick Correlation texture feature for the segmentation of optic disc in colored fundus retinal images. The grayscale and green channel of the colored fundus image are investigated. When compared with some methods in the literature, the experimental study of the proposed method achieved very high average accuracy rates of 98.59% and 98.36% using grayscale and green channel of the colored fundus image respectively on DRIVE database.
{"title":"Optic Disc Segmentation Based on Correlation Feature Information","authors":"Olawuyi O. Fatoki, S. Ojo, P. Owolawi, T. Mapayi","doi":"10.1109/ICONIC.2018.8601255","DOIUrl":"https://doi.org/10.1109/ICONIC.2018.8601255","url":null,"abstract":"Glaucoma, one of the major causes of blindness, has been identified as a disease that causes the degeneration of the optic disc. An highly accurate automated detection of the optic disc (OD) has however been noted to be of great importance for the efficient diagnosis of the disease. This paper presents a study on an unsupervised approach usage of Haralick Correlation texture feature for the segmentation of optic disc in colored fundus retinal images. The grayscale and green channel of the colored fundus image are investigated. When compared with some methods in the literature, the experimental study of the proposed method achieved very high average accuracy rates of 98.59% and 98.36% using grayscale and green channel of the colored fundus image respectively on DRIVE database.","PeriodicalId":277315,"journal":{"name":"2018 International Conference on Intelligent and Innovative Computing Applications (ICONIC)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126484298","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 : 2018-12-01DOI: 10.1109/ICONIC.2018.8601282
Chitra Nundlall, Gopal Sohun, S. Nagowah
Recommender systems are engines that recommend new items to users by analyzing their preferences. The web contains a large amount of information in the form of ratings, reviews, feedback on items and other unstructured data. These details are extracted to get meaningful information of users. Collaborative filtering and content-based filtering are two common approaches being used to make recommendations. The paper aims to introduce a hybrid recommendation technique for Big Data Systems. The approach combines collaborative and content-based filtering techniques to recommend items that a user would most likely prefer. It additionally uses items ranking and classification technique for recommending the items. Moreover, social media opinion mining is added as a top-up to derive user sentiments from user’s posts and become knowledgeable about users’ tastes hidden within social media. A prototype has been implemented and evaluated based on the recommendation techniques.
{"title":"A Hybrid Recommendation Technique for Big Data Systems","authors":"Chitra Nundlall, Gopal Sohun, S. Nagowah","doi":"10.1109/ICONIC.2018.8601282","DOIUrl":"https://doi.org/10.1109/ICONIC.2018.8601282","url":null,"abstract":"Recommender systems are engines that recommend new items to users by analyzing their preferences. The web contains a large amount of information in the form of ratings, reviews, feedback on items and other unstructured data. These details are extracted to get meaningful information of users. Collaborative filtering and content-based filtering are two common approaches being used to make recommendations. The paper aims to introduce a hybrid recommendation technique for Big Data Systems. The approach combines collaborative and content-based filtering techniques to recommend items that a user would most likely prefer. It additionally uses items ranking and classification technique for recommending the items. Moreover, social media opinion mining is added as a top-up to derive user sentiments from user’s posts and become knowledgeable about users’ tastes hidden within social media. A prototype has been implemented and evaluated based on the recommendation techniques.","PeriodicalId":277315,"journal":{"name":"2018 International Conference on Intelligent and Innovative Computing Applications (ICONIC)","volume":"66 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120904651","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 : 2018-12-01DOI: 10.1109/ICONIC.2018.8601254
L. Zulu, K. Ogudo, P. Umenne
In this paper, a Software Defined Network was created in Mininet using python script. An external interface was added in the form of an OpenDaylight controller to enable communication with the network outside of Mininet. The OpenDaylight controller was hosted on the Amazon Web Services elastic computing node. This controller is used as a control plane device for the switch within Mininet. The OpenDaylight controller was able to create the flows to facilitate communication between the hosts in Mininet and the webserver in the real-life network. In order to test the network, a real life network in the form of a webserver hosted on the Emulated Virtual Environment – Next Generation (EVE-NG) software was connected to Mininet.
在本文中，使用python脚本在Mininet中创建了一个软件定义网络。以OpenDaylight控制器的形式添加了一个外部接口，以便与Mininet外部的网络进行通信。OpenDaylight控制器托管在Amazon Web Services弹性计算节点上。该控制器用作Mininet内部交换机的控制平面设备。OpenDaylight控制器能够创建流，以促进Mininet主机与现实网络中的web服务器之间的通信。为了测试网络，一个现实生活中的网络以托管在模拟虚拟环境-下一代(EVE-NG)软件上的web服务器的形式连接到Mininet。
{"title":"Emulating Software Defined Network Using Mininet and OpenDaylight Controller Hosted on Amazon Web Services Cloud Platform to Demonstrate a Realistic Programmable Network","authors":"L. Zulu, K. Ogudo, P. Umenne","doi":"10.1109/ICONIC.2018.8601254","DOIUrl":"https://doi.org/10.1109/ICONIC.2018.8601254","url":null,"abstract":"In this paper, a Software Defined Network was created in Mininet using python script. An external interface was added in the form of an OpenDaylight controller to enable communication with the network outside of Mininet. The OpenDaylight controller was hosted on the Amazon Web Services elastic computing node. This controller is used as a control plane device for the switch within Mininet. The OpenDaylight controller was able to create the flows to facilitate communication between the hosts in Mininet and the webserver in the real-life network. In order to test the network, a real life network in the form of a webserver hosted on the Emulated Virtual Environment – Next Generation (EVE-NG) software was connected to Mininet.","PeriodicalId":277315,"journal":{"name":"2018 International Conference on Intelligent and Innovative Computing Applications (ICONIC)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116597272","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 : 2018-12-01DOI: 10.1109/ICONIC.2018.8601294
Yaseen Khan, S. Thakur
Internet platforms such as Twitter allow cause-related campaigning as well as analysis through the opportunistic classification and aggregation capability provided by the hashtag (#). South African students leveraged Twitter to launch and sustain a campaign now known as the #FeesMustFall campaign. This campaign aimed to lobby government to provide free university education to disadvantaged students. This study examines the #FeesMustFall campaign to determine if automated software robots played a role. The research question was "Did bots and cyborgs play a role in the #FeesMustFall campaign?" 576 823 tweets were harvested, and the data was cleaned by removing duplicate entries. The remaining 490 449 tweets and 90 783 unique users were used to analyze tweet behavior in terms of frequency, volume, content and tweet source. The results show that bots and cyborgs did indeed play a role. This is a significant finding as #FeesMustFall is the first major South African campaign to leverage bots and cyborgs. An important additional finding was the DeBot API revealed 4 bots not found in our harvested tweets while other trait-driven techniques used identified suspicious accounts which revealed two bot or cyborg accounts ranked 1st and 2nd amongst the highest tweeters. This demonstrated a presence of bots during the campaign that assisted in the amplification of the #FeesMustFall hashtag on Twitter.
像Twitter这样的互联网平台允许与事业相关的竞选活动，以及通过标签(#)提供的机会分类和聚合功能进行分析。南非学生利用Twitter发起并维持了一项运动，现在被称为#学费必须下降运动。这项运动旨在游说政府为贫困学生提供免费大学教育。本研究考察了#FeesMustFall运动，以确定自动化软件机器人是否发挥了作用。研究的问题是“机器人和半机器人在#学费必须下降运动中发挥了作用吗?”收集了576 823条tweet，并通过删除重复条目来清理数据。剩余的490449条推文和90783个独立用户从频率、数量、内容和推文来源等方面分析推文行为。结果表明，机器人和半机械人确实发挥了作用。这是一个重要的发现，因为# fee must fall是南非第一个利用机器人和电子人的大型运动。另一个重要的发现是，DeBot API发现了我们收集的推文中没有发现的4个机器人，而使用的其他特征驱动技术发现了可疑账户，其中两个机器人或半机械人账户在推特用户中排名第一和第二。这表明在竞选期间机器人的存在，帮助扩大了推特上的#FeesMustFall标签。
{"title":"The Presence of Twitter Bots and Cyborgs in the #FeesMustFall Campaign","authors":"Yaseen Khan, S. Thakur","doi":"10.1109/ICONIC.2018.8601294","DOIUrl":"https://doi.org/10.1109/ICONIC.2018.8601294","url":null,"abstract":"Internet platforms such as Twitter allow cause-related campaigning as well as analysis through the opportunistic classification and aggregation capability provided by the hashtag (#). South African students leveraged Twitter to launch and sustain a campaign now known as the #FeesMustFall campaign. This campaign aimed to lobby government to provide free university education to disadvantaged students. This study examines the #FeesMustFall campaign to determine if automated software robots played a role. The research question was \"Did bots and cyborgs play a role in the #FeesMustFall campaign?\" 576 823 tweets were harvested, and the data was cleaned by removing duplicate entries. The remaining 490 449 tweets and 90 783 unique users were used to analyze tweet behavior in terms of frequency, volume, content and tweet source. The results show that bots and cyborgs did indeed play a role. This is a significant finding as #FeesMustFall is the first major South African campaign to leverage bots and cyborgs. An important additional finding was the DeBot API revealed 4 bots not found in our harvested tweets while other trait-driven techniques used identified suspicious accounts which revealed two bot or cyborg accounts ranked 1st and 2nd amongst the highest tweeters. This demonstrated a presence of bots during the campaign that assisted in the amplification of the #FeesMustFall hashtag on Twitter.","PeriodicalId":277315,"journal":{"name":"2018 International Conference on Intelligent and Innovative Computing Applications (ICONIC)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114297082","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 : 2018-12-01DOI: 10.1109/ICONIC.2018.8601202
R. Moloo, Kavi Kumar Khedo, T. Prabhakar
This work is a continuation of a larger research work which advocates that Distance Education (DE) through audio-only learning mode can be developed into a full fledge audio-MOOC. Audio MOOC framework is an innovative framework which enables learning through mere phone calls. It has been conceived to digitally include low literate population in the education process by opening up access to learning materials to the unreached and the have-nots usually hindered by barriers such as language, literacy, culture, connectivity and distance which existing MOOCs have failed to address. This work demonstrates how our proposed framework is used to connect to a remote island lost in the middle of the Indian Ocean with limited maritime and air access but which since some few years back can be connected via basic phones through voice calls. Agalega is an ideal test case scenario for our research since it characterizes remoteness, limited connectivity, semi-literate population with limited access to education which our research aims at addressing. A group of 50 Fishermen was identified from both the Agalega islands. The course was of 9 days duration from 15 to 23 September 2017. The system was conducted live over the telephony network making use of our GSM gateway. The specificity of the system was that our GSM gateway resided in Mauritius connected to a cloud server, while the course was delivered to people of Agalega 1100 Km far from Mauritius over the sea. Nevertheless, our system performed as expected and proved to be a success.
{"title":"Delivery of an Interactive Audio Course on Fisheries Law via Dumb Phones: Agalega Island as a Case Study for Testing a Novel Distance Education Platform - The Audio MOOC","authors":"R. Moloo, Kavi Kumar Khedo, T. Prabhakar","doi":"10.1109/ICONIC.2018.8601202","DOIUrl":"https://doi.org/10.1109/ICONIC.2018.8601202","url":null,"abstract":"This work is a continuation of a larger research work which advocates that Distance Education (DE) through audio-only learning mode can be developed into a full fledge audio-MOOC. Audio MOOC framework is an innovative framework which enables learning through mere phone calls. It has been conceived to digitally include low literate population in the education process by opening up access to learning materials to the unreached and the have-nots usually hindered by barriers such as language, literacy, culture, connectivity and distance which existing MOOCs have failed to address. This work demonstrates how our proposed framework is used to connect to a remote island lost in the middle of the Indian Ocean with limited maritime and air access but which since some few years back can be connected via basic phones through voice calls. Agalega is an ideal test case scenario for our research since it characterizes remoteness, limited connectivity, semi-literate population with limited access to education which our research aims at addressing. A group of 50 Fishermen was identified from both the Agalega islands. The course was of 9 days duration from 15 to 23 September 2017. The system was conducted live over the telephony network making use of our GSM gateway. The specificity of the system was that our GSM gateway resided in Mauritius connected to a cloud server, while the course was delivered to people of Agalega 1100 Km far from Mauritius over the sea. Nevertheless, our system performed as expected and proved to be a success.","PeriodicalId":277315,"journal":{"name":"2018 International Conference on Intelligent and Innovative Computing Applications (ICONIC)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124932461","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 : 2018-12-01DOI: 10.1109/ICONIC.2018.8601243
Mduduzi Manana, Chunling Tu, P. Owolawi
This paper presents a pre-processed faster region convolution neural network (faster RCNN) for the purpose of on-road vehicle detection. The system introduces a preprocessing pipeline on faster RCNN. The preprocessing method is for the improvement on training and detection speed of Faster RCNN. A preprocessing lane detection pipeline based on the Sobel edge operator and Hough Transform is used to detect lanes. A Rectangular region is then extracted from lane coordinates which is a reduced region of interest (ROI). Results show that the proposed method improves the training speed of faster RCNN when compared to faster RCNN without preprocessing.
{"title":"Preprocessed Faster RCNN for Vehicle Detection","authors":"Mduduzi Manana, Chunling Tu, P. Owolawi","doi":"10.1109/ICONIC.2018.8601243","DOIUrl":"https://doi.org/10.1109/ICONIC.2018.8601243","url":null,"abstract":"This paper presents a pre-processed faster region convolution neural network (faster RCNN) for the purpose of on-road vehicle detection. The system introduces a preprocessing pipeline on faster RCNN. The preprocessing method is for the improvement on training and detection speed of Faster RCNN. A preprocessing lane detection pipeline based on the Sobel edge operator and Hough Transform is used to detect lanes. A Rectangular region is then extracted from lane coordinates which is a reduced region of interest (ROI). Results show that the proposed method improves the training speed of faster RCNN when compared to faster RCNN without preprocessing.","PeriodicalId":277315,"journal":{"name":"2018 International Conference on Intelligent and Innovative Computing Applications (ICONIC)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126536260","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 : 2018-12-01DOI: 10.1109/ICONIC.2018.8601258
Mariapaola Saponaro, Diane Le Gal, Manjiao Gao, Matthieu Guisiano, I. Manière
This article is depicting the Strengths and weaknesses of Artificial Intelligence related to the improvement of customer online and offline experience, and the possible methods in order to measure them. These methods include both researches non-based and based on interviews. The presence of AI in the retail industry is becoming a key component of the customer experience. Through a deep analysis of existing tools to extract information, we try to explain ways to interpret them, in order for companies to create a real usage out of them, either on online or offline retail experience. Hence, with this research, we also want to provide an insight on how this experience could be improved in the future, and how it will most likely be inherent to our daily customer experience.
{"title":"Challenges and Opportunities of Artificial Intelligence in the Fashion World","authors":"Mariapaola Saponaro, Diane Le Gal, Manjiao Gao, Matthieu Guisiano, I. Manière","doi":"10.1109/ICONIC.2018.8601258","DOIUrl":"https://doi.org/10.1109/ICONIC.2018.8601258","url":null,"abstract":"This article is depicting the Strengths and weaknesses of Artificial Intelligence related to the improvement of customer online and offline experience, and the possible methods in order to measure them. These methods include both researches non-based and based on interviews. The presence of AI in the retail industry is becoming a key component of the customer experience. Through a deep analysis of existing tools to extract information, we try to explain ways to interpret them, in order for companies to create a real usage out of them, either on online or offline retail experience. Hence, with this research, we also want to provide an insight on how this experience could be improved in the future, and how it will most likely be inherent to our daily customer experience.","PeriodicalId":277315,"journal":{"name":"2018 International Conference on Intelligent and Innovative Computing Applications (ICONIC)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130005871","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 : 2018-12-01DOI: 10.1109/ICONIC.2018.8601096
Fawaz S. Al-Anzi, Dia AbuZeina
The Vector Space Model (VSM) is a common document representation model that is widely used in data mining and information retrieval (IR) systems. However, this technique poses some challenges such as high dimensional space and semantic loss representation. Therefore, the latent semantic indexing (LSI) is proposed to reduce the feature dimensions and to generate semantic rich features that represent conceptual term-document associations. In particular, LSI has been successfully implemented in search engines and text classification tasks. In this paper, we propose a novel approach to enhance the quality of the retrieved documents in search engines for Arabic language. That is, we propose to use a new extension of the LSI technique instead of just using the standard LSI technique. The LSI method proposed is based on employing the word co-occurrences to form a term-by-document matrix. The proposed method is to be based on the documents evaluating cosine similarity measures for term-by-document matrix. We will empirically evaluate the performance using an Arabic data collection that contains no less than 500 documents with no less than 30,000 unique words. A testing set contains keywords from a specific domain will be used to evaluate the quality of the top 20-30 retrieved documents using different singular values (i.e. different number of dimensions). The results will be judged on the performance of the proposed method as it is compared to the standard LSI.
{"title":"Enhanced Search for Arabic Language Using Latent Semantic Indexing (LSI)","authors":"Fawaz S. Al-Anzi, Dia AbuZeina","doi":"10.1109/ICONIC.2018.8601096","DOIUrl":"https://doi.org/10.1109/ICONIC.2018.8601096","url":null,"abstract":"The Vector Space Model (VSM) is a common document representation model that is widely used in data mining and information retrieval (IR) systems. However, this technique poses some challenges such as high dimensional space and semantic loss representation. Therefore, the latent semantic indexing (LSI) is proposed to reduce the feature dimensions and to generate semantic rich features that represent conceptual term-document associations. In particular, LSI has been successfully implemented in search engines and text classification tasks. In this paper, we propose a novel approach to enhance the quality of the retrieved documents in search engines for Arabic language. That is, we propose to use a new extension of the LSI technique instead of just using the standard LSI technique. The LSI method proposed is based on employing the word co-occurrences to form a term-by-document matrix. The proposed method is to be based on the documents evaluating cosine similarity measures for term-by-document matrix. We will empirically evaluate the performance using an Arabic data collection that contains no less than 500 documents with no less than 30,000 unique words. A testing set contains keywords from a specific domain will be used to evaluate the quality of the top 20-30 retrieved documents using different singular values (i.e. different number of dimensions). The results will be judged on the performance of the proposed method as it is compared to the standard LSI.","PeriodicalId":277315,"journal":{"name":"2018 International Conference on Intelligent and Innovative Computing Applications (ICONIC)","volume":"126 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128063159","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：