Online ISSN: 2515-8260

Author : Ashesh, K.


K. Ashesh; Dr.G. AppaRao

European Journal of Molecular & Clinical Medicine, 2020, Volume 7, Issue 4, Pages 1078-1084

As the name specifies “HOSTEL MANAGEMENT SYSTEM” is a project which is developed for the management of several activities that happens in every hostel. These days there are various educational institutions increasing for every day and these are followed by the requirement of accommodation for students who wanted to study in these institutions. So there is a lot of work to be done for a student to be joined in this hostel and in these contexts generally there are no applications or any software developed. So our project is much helpful in these contexts. This project is developed based on problems faced by the hostel management and avoids many problems which are caused by man work which causes problems like incorrect data entry and several other problems. By identifying these drawbacks caused by current technique which interrelated for developing an automated and digitalized system known to be the hostel management system which gives much more comfortable and compatible with the existing system and this is the system which is more user friendly. The proposed system has superior efficiency, and it surmounts the shortcomings of existing technique of hotel management system. Also, it has few advantages while the system is generating such as providing high security, avoiding data redundancy, data consistency, easy handling, stored record, data updating, less laborious work and less human error.

D3O: A Framework for Distributed Distance-based Detection of Outliers in Large Data Sets

K. Ashesh; Dr.G. AppaRao

European Journal of Molecular & Clinical Medicine, 2020, Volume 7, Issue 4, Pages 1064-1077

Data comes from diversified sources in a distributed computing environment. Outlier detection in such environment is challenging as it involves a strategy to mine outliers. Parallel processing of data available in multiple sources can provide outliers in short span of time. In fact speed with which outlier are mined and interpreted to make well informed decisions is very important in many real world applications like disease outburst detection in healthcare domain. Towards this end, in this paper, we proposed a framework known as Distributed Distance-based Detection of Outliers (D30). The framework guides the process of discovering outliers from large data sets. An algorithm named Distributed Outlier Detection (DOD) is proposed to achieve this. The algorithm exploits the notion of outlier detection solving set to have effective detection of outliers. Two synthetic datasets known as G2d and G3d and a real dataset from NASA named 2Mass are used to evaluate the proposed algorithm. We built a prototype application to demonstrate proof of the concept. The empirical results revealed that the proposed algorithm is capable of finding outliers effectively. The algorithm showed better performance when compared with other state of the art outlier detection algorithm that employs distributed approach in mining outliers