Pub Date : 2020-10-09DOI: 10.1109/ICSTCEE49637.2020.9277236
Mukundan Sundararajan, Siddharth K. Saraya, Sankar Ghoshal, Balaji Ramaswamy
Groups’ features in digital tools provide easy and simple ways to communicate with multiple members who share a common objective or participate together in activities to achieve certain shared goals. Number of groups in enterprises, institutions and social networks has grown tremendously. Groups have their identity and a membership list in the digital spaces that need to be managed efficiently. Addition of new members to groups or changing roles within a group has found a disciplined approach and adoption. However pruning of groups based on activity and participation has not been a strength in group management in most organizations which impacts information security and time of members in the groups from unnecessary calendar entries due to this sub-optimal group management. The paper analyzes impacts from efforts involved with the corrective measures undertaken today due to the deficiencies and proposes measures and steps to improve the group management activities to reduce the impacts. The paper discusses potential automation or feature enhancements in the digital tools that enables measurement to trigger improvements in the methods of groups’ management to mitigate the said impacts.
{"title":"Managing dynamic group membership in the evolving digital spaces","authors":"Mukundan Sundararajan, Siddharth K. Saraya, Sankar Ghoshal, Balaji Ramaswamy","doi":"10.1109/ICSTCEE49637.2020.9277236","DOIUrl":"https://doi.org/10.1109/ICSTCEE49637.2020.9277236","url":null,"abstract":"Groups’ features in digital tools provide easy and simple ways to communicate with multiple members who share a common objective or participate together in activities to achieve certain shared goals. Number of groups in enterprises, institutions and social networks has grown tremendously. Groups have their identity and a membership list in the digital spaces that need to be managed efficiently. Addition of new members to groups or changing roles within a group has found a disciplined approach and adoption. However pruning of groups based on activity and participation has not been a strength in group management in most organizations which impacts information security and time of members in the groups from unnecessary calendar entries due to this sub-optimal group management. The paper analyzes impacts from efforts involved with the corrective measures undertaken today due to the deficiencies and proposes measures and steps to improve the group management activities to reduce the impacts. The paper discusses potential automation or feature enhancements in the digital tools that enables measurement to trigger improvements in the methods of groups’ management to mitigate the said impacts.","PeriodicalId":113845,"journal":{"name":"2020 International Conference on Smart Technologies in Computing, Electrical and Electronics (ICSTCEE)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123564474","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 : 2020-10-09DOI: 10.1109/ICSTCEE49637.2020.9277256
S. Sindhu, Sunil Parameshwar Patil, Arya Sreevalsan, F. Rahman, Ms. Saritha A. N.
Phishing is a common attack used to obtain sensitive information using visually similar websites to that of legitimate websites. With the growing technology, phishing attacks are on the rise. Machine Learning is a very popular approach to detect phishing websites. This paper explains the existing machine learning methods that are used to detect phishing websites. The paper explains the improved Random Forest classification method, SVM classification algorithm and Neural Network with backpropagation classification methods which have been implemented with accuracies of 97.369%, 97.451% and 97.259% respectively.
{"title":"Phishing Detection using Random Forest, SVM and Neural Network with Backpropagation","authors":"S. Sindhu, Sunil Parameshwar Patil, Arya Sreevalsan, F. Rahman, Ms. Saritha A. N.","doi":"10.1109/ICSTCEE49637.2020.9277256","DOIUrl":"https://doi.org/10.1109/ICSTCEE49637.2020.9277256","url":null,"abstract":"Phishing is a common attack used to obtain sensitive information using visually similar websites to that of legitimate websites. With the growing technology, phishing attacks are on the rise. Machine Learning is a very popular approach to detect phishing websites. This paper explains the existing machine learning methods that are used to detect phishing websites. The paper explains the improved Random Forest classification method, SVM classification algorithm and Neural Network with backpropagation classification methods which have been implemented with accuracies of 97.369%, 97.451% and 97.259% respectively.","PeriodicalId":113845,"journal":{"name":"2020 International Conference on Smart Technologies in Computing, Electrical and Electronics (ICSTCEE)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125667099","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 : 2020-10-09DOI: 10.1109/ICSTCEE49637.2020.9276787
Sandeep Kumar, K. Prasad, M. Manish Gupta, B. Pavani, C. Deepak Reddy, John Moses
In the recent years digital home assistance is expanding rapidly because of its huge demands in the market. As the technology is getting evolved day by day, the need to access to things in a smarter way is getting stronger. The smart bulb, smart house, smart cities, smart watches, smart phones all the regularly used devices in our daily lives is getting smarter in a blink of an eye. The latest device that has been built is the VIRTUAL ASSISTANT. JOYO is the house assistant that can assist you in your daily life. JOYO is built on the latest technology Artificial Intelligence and the Internet of Things. Apart from assisting you, JOYO can actually control the things for you such as the indoor appliances as well as outdoor appliances. JOYO can talk to you, play music for you, turn on or off appliances, move according to the directions given, and all these works are done just by listening to your command. Call JOYO and JOYO wakes up, call JOYO TURN OFF JOYO will turn OFF.
{"title":"Joyo: The House Assistant Technology for Smart Home","authors":"Sandeep Kumar, K. Prasad, M. Manish Gupta, B. Pavani, C. Deepak Reddy, John Moses","doi":"10.1109/ICSTCEE49637.2020.9276787","DOIUrl":"https://doi.org/10.1109/ICSTCEE49637.2020.9276787","url":null,"abstract":"In the recent years digital home assistance is expanding rapidly because of its huge demands in the market. As the technology is getting evolved day by day, the need to access to things in a smarter way is getting stronger. The smart bulb, smart house, smart cities, smart watches, smart phones all the regularly used devices in our daily lives is getting smarter in a blink of an eye. The latest device that has been built is the VIRTUAL ASSISTANT. JOYO is the house assistant that can assist you in your daily life. JOYO is built on the latest technology Artificial Intelligence and the Internet of Things. Apart from assisting you, JOYO can actually control the things for you such as the indoor appliances as well as outdoor appliances. JOYO can talk to you, play music for you, turn on or off appliances, move according to the directions given, and all these works are done just by listening to your command. Call JOYO and JOYO wakes up, call JOYO TURN OFF JOYO will turn OFF.","PeriodicalId":113845,"journal":{"name":"2020 International Conference on Smart Technologies in Computing, Electrical and Electronics (ICSTCEE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130790204","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 : 2020-10-09DOI: 10.1109/icstcee49637.2020.9276939
{"title":"[ICSTCEE 2020 Front matter]","authors":"","doi":"10.1109/icstcee49637.2020.9276939","DOIUrl":"https://doi.org/10.1109/icstcee49637.2020.9276939","url":null,"abstract":"","PeriodicalId":113845,"journal":{"name":"2020 International Conference on Smart Technologies in Computing, Electrical and Electronics (ICSTCEE)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134267959","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 : 2020-10-09DOI: 10.1109/ICSTCEE49637.2020.9277446
R. R, S. Saravanan, Gopal Krishna Shyam
In this paper, we reveal the correlation between two seemingly unrelated topics which is BMI and Mental Illness using a Machine Learning approach. Body mass plays an important part for mental illness e.g., Depression and anxiety .It deals with a person’s mass and height. A person's weight has a drastic effect on their lifestyle and health. Large valued BMI's are linked to all kinds of diseases ranging from diabetes to heart disease. Mental disorders and obesity are chronic conditions which need attention and care. This paper compares the relationship BMI has to mental diseases. We have used Machine learning algorithm to solve the problem concerned. To find the relationship between the features of our dataset, we performed Linear Regression. Here we tried to find the relationship between BMI and Mental Illness, depression more specifically. We also observed experimentally that the risk of a person who was overweight, obese or extremely obese to develop a psychiatric illness were 45-90 per cent higher than a person of average weight. We received an accuracy value of 0.690, after applying Linear Regression to the dataset. Using more sophisticated machine learning techniques would increase the precision. Experimental findings indicate that the method presented is almost equivalent to other state-of-the-art models.
{"title":"The BMI and Mental Illness Nexus: A Machine Learning Approach","authors":"R. R, S. Saravanan, Gopal Krishna Shyam","doi":"10.1109/ICSTCEE49637.2020.9277446","DOIUrl":"https://doi.org/10.1109/ICSTCEE49637.2020.9277446","url":null,"abstract":"In this paper, we reveal the correlation between two seemingly unrelated topics which is BMI and Mental Illness using a Machine Learning approach. Body mass plays an important part for mental illness e.g., Depression and anxiety .It deals with a person’s mass and height. A person's weight has a drastic effect on their lifestyle and health. Large valued BMI's are linked to all kinds of diseases ranging from diabetes to heart disease. Mental disorders and obesity are chronic conditions which need attention and care. This paper compares the relationship BMI has to mental diseases. We have used Machine learning algorithm to solve the problem concerned. To find the relationship between the features of our dataset, we performed Linear Regression. Here we tried to find the relationship between BMI and Mental Illness, depression more specifically. We also observed experimentally that the risk of a person who was overweight, obese or extremely obese to develop a psychiatric illness were 45-90 per cent higher than a person of average weight. We received an accuracy value of 0.690, after applying Linear Regression to the dataset. Using more sophisticated machine learning techniques would increase the precision. Experimental findings indicate that the method presented is almost equivalent to other state-of-the-art models.","PeriodicalId":113845,"journal":{"name":"2020 International Conference on Smart Technologies in Computing, Electrical and Electronics (ICSTCEE)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133121024","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 : 2020-10-09DOI: 10.1109/ICSTCEE49637.2020.9276897
Rajesh Rompicharla, B. P. V.
Cloud Computing offers instant Infrastructure for the needs of Application & Development teams, software development phase accelerated with the adaption of DevOps model. DevOps model key tenets of success are Self-Service, Permission to act, Guardrails and Trust. Since the current Cloud Computing trend is moving towards Hybrid Multi-Cloud the Multi-Tenant phenomenon as well geared up to meet the agile business needs. The Security team's strategy for DevOps, especially related to its Self-Service tenet needs an immediate review. According to the survey in 2019, around 90% of enterprises use some type of Cloud Service; around 50% already adapted Hybrid Multi-Cloud; still 67% security teams had lack of visibility into their cloud infrastructure, security and compliance. Attacks due to Misconfigured Cloud environments was the main cause of data theft Security Incidents. Popular Incidents related to Amazon about Facebook accounts leak and Microsoft about data theft related to Box accounts primary cause is misconfiguration of the concern parties. Hence, Post-Deployment Compliance checks does not suffice the needs of required Security for Hybrid Multi-Cloud environments. Continuous and Pre-Deployment Compliant Self-Service solution for Hybrid Multi-Cloud with appropriate design and implementation procedure is the objective of this paper.
{"title":"Continuous Compliance model for Hybrid Multi-Cloud through Self-Service Orchestrator","authors":"Rajesh Rompicharla, B. P. V.","doi":"10.1109/ICSTCEE49637.2020.9276897","DOIUrl":"https://doi.org/10.1109/ICSTCEE49637.2020.9276897","url":null,"abstract":"Cloud Computing offers instant Infrastructure for the needs of Application & Development teams, software development phase accelerated with the adaption of DevOps model. DevOps model key tenets of success are Self-Service, Permission to act, Guardrails and Trust. Since the current Cloud Computing trend is moving towards Hybrid Multi-Cloud the Multi-Tenant phenomenon as well geared up to meet the agile business needs. The Security team's strategy for DevOps, especially related to its Self-Service tenet needs an immediate review. According to the survey in 2019, around 90% of enterprises use some type of Cloud Service; around 50% already adapted Hybrid Multi-Cloud; still 67% security teams had lack of visibility into their cloud infrastructure, security and compliance. Attacks due to Misconfigured Cloud environments was the main cause of data theft Security Incidents. Popular Incidents related to Amazon about Facebook accounts leak and Microsoft about data theft related to Box accounts primary cause is misconfiguration of the concern parties. Hence, Post-Deployment Compliance checks does not suffice the needs of required Security for Hybrid Multi-Cloud environments. Continuous and Pre-Deployment Compliant Self-Service solution for Hybrid Multi-Cloud with appropriate design and implementation procedure is the objective of this paper.","PeriodicalId":113845,"journal":{"name":"2020 International Conference on Smart Technologies in Computing, Electrical and Electronics (ICSTCEE)","volume":"128 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114507369","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 : 2020-10-09DOI: 10.1109/ICSTCEE49637.2020.9277088
T. M. Murugan, E. Baburaj
K-medoids clustering aims at partitioning a numerous data points into different K medoids based on the similarity distance between object pair. Cluster efficiency varies with respect to medoid initialization and may results in local optima traps. Various evolutionary swarm based approaches are adopted to obtain enhanced performance. Suitable combination of K-medoids with optimization technique does not operate well as expected while considering computational time. No such techniques are developed to solve entire clustering drawbacks. Assurance is not provided to any method in attaining success by solving the overall issues. Hence in this, an improved K-medoids integrated with Ant lion optimization and Particle swarm optimization algorithm commonly referred as ALPSOC is proposed to obtain optimized cluster centroid in which the computational complexity is preserved with better improvements in performance. Further the intra-cluster distance, F-measure, Rand Index, Adjusted Rand Index, Entropy and Normalized Mutual Information is evaluated for different datasets adopting the presented approach. The proposed algorithm is simulated on different datasets and is compared with different existing techniques based on above performance metric. From the observed results, it is shown that the proposed method functions better in all cases maintaining to solve clustering limitations.
{"title":"Alpsoc Ant Lion * : Particle Swarm Optimized Hybrid K-Medoid Clustering","authors":"T. M. Murugan, E. Baburaj","doi":"10.1109/ICSTCEE49637.2020.9277088","DOIUrl":"https://doi.org/10.1109/ICSTCEE49637.2020.9277088","url":null,"abstract":"K-medoids clustering aims at partitioning a numerous data points into different K medoids based on the similarity distance between object pair. Cluster efficiency varies with respect to medoid initialization and may results in local optima traps. Various evolutionary swarm based approaches are adopted to obtain enhanced performance. Suitable combination of K-medoids with optimization technique does not operate well as expected while considering computational time. No such techniques are developed to solve entire clustering drawbacks. Assurance is not provided to any method in attaining success by solving the overall issues. Hence in this, an improved K-medoids integrated with Ant lion optimization and Particle swarm optimization algorithm commonly referred as ALPSOC is proposed to obtain optimized cluster centroid in which the computational complexity is preserved with better improvements in performance. Further the intra-cluster distance, F-measure, Rand Index, Adjusted Rand Index, Entropy and Normalized Mutual Information is evaluated for different datasets adopting the presented approach. The proposed algorithm is simulated on different datasets and is compared with different existing techniques based on above performance metric. From the observed results, it is shown that the proposed method functions better in all cases maintaining to solve clustering limitations.","PeriodicalId":113845,"journal":{"name":"2020 International Conference on Smart Technologies in Computing, Electrical and Electronics (ICSTCEE)","volume":"214 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116427172","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 : 2020-10-09DOI: 10.1109/ICSTCEE49637.2020.9276957
I. Preethi, K. Dharmarajan
Today, digitization in healthcare industry takes the advantage on advancements in clinical healthcare services. The extensive growth in data for monitoring and analyzing the patients outcomes in predicting and diagnosis of chronic diseases lacks in traditional methods and are replaced by technologies to gather the most relevant insights from the medical data by using predictive analytics with very useful tool of machine learning. The importance of using machine learning algorithms in the model for diagnosis, shows its ability in high classification accuracy rate in reduced computational time. In this paper, a study of various machine learning techniques are used in classification of chronic diseases like heart, kidney, diabetes and cancer from multiple dataset by reducing the dimensionality using feature selection. Feature selection plays a significant role in machine learning by selecting the critical features for diagnosing chronic diseases. The performance of the classifiers are evaluated based on several metrics like classification accuracy, sensitivity, specificity, precision, F1- measure, AUC (the area under the receiver operating characteristic (ROC) curve) criterion, and processing time.
{"title":"Diagnosis of chronic disease in a predictive model using machine learning algorithm","authors":"I. Preethi, K. Dharmarajan","doi":"10.1109/ICSTCEE49637.2020.9276957","DOIUrl":"https://doi.org/10.1109/ICSTCEE49637.2020.9276957","url":null,"abstract":"Today, digitization in healthcare industry takes the advantage on advancements in clinical healthcare services. The extensive growth in data for monitoring and analyzing the patients outcomes in predicting and diagnosis of chronic diseases lacks in traditional methods and are replaced by technologies to gather the most relevant insights from the medical data by using predictive analytics with very useful tool of machine learning. The importance of using machine learning algorithms in the model for diagnosis, shows its ability in high classification accuracy rate in reduced computational time. In this paper, a study of various machine learning techniques are used in classification of chronic diseases like heart, kidney, diabetes and cancer from multiple dataset by reducing the dimensionality using feature selection. Feature selection plays a significant role in machine learning by selecting the critical features for diagnosing chronic diseases. The performance of the classifiers are evaluated based on several metrics like classification accuracy, sensitivity, specificity, precision, F1- measure, AUC (the area under the receiver operating characteristic (ROC) curve) criterion, and processing time.","PeriodicalId":113845,"journal":{"name":"2020 International Conference on Smart Technologies in Computing, Electrical and Electronics (ICSTCEE)","volume":"542 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132903769","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 : 2020-10-09DOI: 10.1109/ICSTCEE49637.2020.9277417
Archit Kapoor, Divyansh Oze, A. Shankar
Internet of Things home automation industry is opening the flood gates to a whole new world of sensors and microcontrollers that reduce human effort and simplify how usual things function around us. This project aims to explore ESP8266 and its capabilities along with PIR sensors to make a home automation prototype that focusses on "Conservation of Energy". By establishing a Web Server which would record the data of the number of persons entering/exiting a given room, we explore the ability of this module to use wifi protocols, being a microcontroller. The setup process for the apparatus requires feasibility, perseverance and precision. Once accomplished we move on to the equally difficult process of setting up an Integrated Development Environment (IDE) and pushing our program for the server onto the moduleKeeping in mind the potential of the project, the future scope including applications using the project has been discussed in this project.
{"title":"IoT Aided Smart Light Sensing Automation using Passive Infrared Sensors","authors":"Archit Kapoor, Divyansh Oze, A. Shankar","doi":"10.1109/ICSTCEE49637.2020.9277417","DOIUrl":"https://doi.org/10.1109/ICSTCEE49637.2020.9277417","url":null,"abstract":"Internet of Things home automation industry is opening the flood gates to a whole new world of sensors and microcontrollers that reduce human effort and simplify how usual things function around us. This project aims to explore ESP8266 and its capabilities along with PIR sensors to make a home automation prototype that focusses on \"Conservation of Energy\". By establishing a Web Server which would record the data of the number of persons entering/exiting a given room, we explore the ability of this module to use wifi protocols, being a microcontroller. The setup process for the apparatus requires feasibility, perseverance and precision. Once accomplished we move on to the equally difficult process of setting up an Integrated Development Environment (IDE) and pushing our program for the server onto the moduleKeeping in mind the potential of the project, the future scope including applications using the project has been discussed in this project.","PeriodicalId":113845,"journal":{"name":"2020 International Conference on Smart Technologies in Computing, Electrical and Electronics (ICSTCEE)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132979482","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 : 2020-10-09DOI: 10.1109/ICSTCEE49637.2020.9277129
C. Rashmi, M. Kodabagi
The era of today’s world is based on multiple online social network such as twitter, Facebook, LinkedIn and many more. User’s use this online social network for commination in terms of text, audio, video, images, gif’s and so on which leads to enormous amount of unstructured data generation. Hence, it becomes inevitable to analyze these unstructured data into meaningful knowledge which can be applied to various applications such as link prediction, criminology, public health, recommendation system and many more. Most applications of social networks require user profile data to analyze the data. In this paper, we propose a graph based methodology to connect user profiles based on their attributes similarity and build a social network graph of connected users. The methodology is tested on LinkedIn data set and results are promising. The methodology addresses various issues associated with unstructured data analysis.
{"title":"Profiling of Social Network Users using Proximity Measures","authors":"C. Rashmi, M. Kodabagi","doi":"10.1109/ICSTCEE49637.2020.9277129","DOIUrl":"https://doi.org/10.1109/ICSTCEE49637.2020.9277129","url":null,"abstract":"The era of today’s world is based on multiple online social network such as twitter, Facebook, LinkedIn and many more. User’s use this online social network for commination in terms of text, audio, video, images, gif’s and so on which leads to enormous amount of unstructured data generation. Hence, it becomes inevitable to analyze these unstructured data into meaningful knowledge which can be applied to various applications such as link prediction, criminology, public health, recommendation system and many more. Most applications of social networks require user profile data to analyze the data. In this paper, we propose a graph based methodology to connect user profiles based on their attributes similarity and build a social network graph of connected users. The methodology is tested on LinkedIn data set and results are promising. The methodology addresses various issues associated with unstructured data analysis.","PeriodicalId":113845,"journal":{"name":"2020 International Conference on Smart Technologies in Computing, Electrical and Electronics (ICSTCEE)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121784128","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：