International Journal of Social Computing and Cyber-Physical Systems (6 papers in press)
Implementation of an Efficient and Intelligent Indian Maritime Borderline Alert System using IoT
by Asha Jerlin, Anindya Kumar Das
Abstract: The sea border issues are crucial to nations where its hard to have a track of boarder for fishermens. Quite a few occurrences of Indian fishermen from Tamilnadu, have been facing problems near the historic maritime boundary. About 180000 fishing vessels of assorted forms are involved in fishing along the Indian-Sri Lankan maritime border. But fishing activities in that region has always been an agitating factor, ever since violence broke out in Sri Lanka about 3 decades ago. This problem however this can be solved only by an efficient and intelligent boundary alert system using IoT. Internet of Things has the technology to make it possible for monitoring the sea border levels to know the border location. The Intelligent Maritime Borderline Alter System or IMBAS incorporates the new methodology of IoT for saving the fishermens life as well as their sole assets in which their mode of income depends. The methodology includes tracking the position of the boat in real time using a differential GPS system and keeping the boat in safe waters, well inside the Indian Territory. The system will also notify the ground control as and when the boat approaches the maritime borderline. The main objective of the system is to help the fishermen navigate safely inside our maritime country border and also prevent them from crossing it at all costs.
Keywords: maritime borderline; India-Sri Lanka; GPS; border security; vessel monitoring; alert system; Tamil Nadu; fishing community; IoT.
A Novel Approach to Text Clustering using Shift K-Medoid
by Mohit , A. Charan Kumari, Meghna Sharma
Abstract: As the amount of data is growing day by day, we need to convert it into some effective manner so as to extract some useful information from huge data. Text Mining is used to perform this task. We use text clustering to convert the large data into different cluster forms to extract the meaningful information for the purpose of analysis so as to get the summarized data. Three partitioning-based clustering techniques i.e. k-means, k-means fast and k-medoids are compared, and a new algorithm named shift k-medoid is proposed, which is hybrid of k-medoid and mean shift clustering algorithms. Cosine Similarity, Correlation Coefficient and Jaccard Similarity measures are used to check the performance of the algorithms and two measures i.e. Randomized feature and Normalized Mutual Information (NMI) feature are used to test the accuracy of the algorithms. The outcomes demonstrate that the best performance is accomplished by using our proposed algorithm.
Keywords: Text Clustering; Cosine measure; Jaccard measure; Correlation coefficient; Shift k-medoid.
The troika of Artificial Intelligence, Emotional Intelligence and Customer Intelligence
by Manish Sharma, Shikha Khera, Pritam B. Sharma
Abstract: Abstract: Emotional Intelligence is to recognise emotions and emotions can be recognised by analysing face. Face reflects emotions, and thus facial images can help to identify emotions. Emotions recognition can help in conducting qualitative market research techniques like focus groups; in-depth interviews and other which can be used to generate customer intelligence. This paper provides a cross-disciplinary view of Intelligence. This paper proposes a machine learning based model to accomplish the task of identifying emotions from given facial images. This paper uses a public database and divides the images into four groups. The feature extraction has been done by Principal Component Analysis and the feature selection by Fisher Discriminant ratio. The classification has been done by Support Vector Machine using k-cross-validation. The accuracy, specificity and sensitivity are encouraging. The average accuracy is 0.84
Keywords: Keywords: Emotions;Emotional Intelligence; Customer Intelligence; Support Vector Machine; Principal Component Analysis; Cross Validation; Machine Learning.
Privacy-preserving Targeted Online Advertising
by Ainish Dave, Hardik Gajera, Maniklal Das
Abstract: Preserving privacy of the user data while availing online services is one of the important requirements as well as challenging tasks, as the service provider can track the user, user's activity, misuse of user data by sharing the data with online advertisement companies, and so on. Out of these many threats that cause user's privacy for online services, the targeted advertisement is found a potential business model that the services provider can use it for increasing revenue or sell it off to other service providers. Typically, the advertising server (ad-server) tracks the users activity through the user's browsing history or through the user's mobile application feature for future purposes. As a result, the privacy of user's data could fall in hand of many parties without the knowledge (and consent) of the user. In this paper, a scheme for privacy-preserving targeted online advertisement is presented. The proposed scheme uses homomorphic encryption aiming at extracting keywords from web pages in real-time when a user is browsing web pages. With the proposed scheme a user can get the intended services privately from the ad-server, which can perform the computation on encrypted data stored in the server. The proposed scheme is efficient with respect to computational cost and in terms of the size of data transfer between client and sever. The proposed scheme is compared with some related schemes and the experimental results show that the proposed scheme is practical for implementation in real-world applications.
Keywords: Privacy; Private information retrieval; Targeted advertising; Homomorphic encryption.
Cyber-Squatting: A Cyber Crime more than an Unethical Act
by Ramesh C. Poonia, Vaibhav Bhatnagar
Abstract: Cyber-crimes are increasing day by day. They equally affect a person or a
society as compared to other crimes. There are so many cyber-crimes such as hacking,
phishing and cyber terrorism etc. However there are some unethical activities that also
affect a person or society such as cyber-squatting. This paper gives a glance about cyber-squatting which is consider as unethical practice, but as per the rate it is increasing day by day it is becoming necessary that it should be included in constitution. This paper illustrates cyber-squatting with the help of examples and some case studies.
Keywords: Cyber-Squatting; Typo Squatting; Domain Squatting;.
Authentication Schemes for Social Network Users: A Review
by Venkataram Pallapa, Swapnil Ninawe
Abstract: Social networks have been very successful for information sharing due to the rapid development of communication and networking. As the technology modernised and many unprotected types of equipment are used in sharing information like events, knowledge, activities, etc., there is a need for a high protection of social networking sites by providing dynamic authentication for social networks users. It is no longer a question of determining whether a user is who or what he/she is declared to be, but it is essential to design a convenient and credible authentication based on activity, relation, etc., of users. Managing and handling such authentication of users is complex and hard for which broad range of technologies need to be called. In this paper, we review some of the existing authentication schemes for social network users. Mainly, we highlight advantages of the existing schemes along with some of the improvements that can be incorporated. We also highlight some of the attacks and their prevention in social networks.
Keywords: Authentication; User; Relation; Social Networks.