Informatics Publishing Limited

L. B. Krithika ¹, Anand Mahendran ², A. Kalyana Vasanth ¹, G. R. Reddy Harish ¹, T. Hariharan ¹

Affiliations
1 School of Information and Technology and Engineering, IN
2 School of Computing Science and Engineering, VIT University, Vellore, Tamilnadu, IN

Abstract

Natural language processing (NLP) has become a main context for many websites. When we consider the case of social networking websites, many posts and conversations in the present day lacks in punctuation and vocabulary. People use short forms such as wru (where are you), hru (how are you) etc., in messages and comments in social networking sites. The comments with respect to the posts need to be categorized into positive, neutral or negative comments. Suppose a person post a status and people start commenting on it, there should be a measure above the post indicating the percentage of positive, negative and neutral comments. Hence, there is need for developing a tool-tip translator, which expands such short forms when mouse cursor is placed on them. There is also a need for developing a text classification system that classifies text based on polarity. Positive comments will be marked in green colour and negative comments will be marked in red colour and neutral in yellow. This paper shows how to implement such a tooltip translator and text classification system.

Keywords

Natural Language Processing, Opinion Mining, Tooltip Translator, Sentiment Analysis, Social Networking

Full Text

   

Kavin Nagasubramanian , I. Fazil Ahmed , T. Hariharan , S. Surya Prabhakaran , S. Karthiga

Abstract

If the number of user grows in huge amount in a Recommendation System, the standard approach of sequentially examining each item and looking at all interacting users does not scale. In our proposed system we solve this problem by developing a MapReduce algorithm for the item comparison and Top-N recommendation problem that scales linearly with respect to a growing number of users. We use Similarity-based neighborhood methods for recommendation; infer their predictions by finding users with similar taste or items that have been similarly rated. In Mapreduce, the data to process is split and stored block-wise across the machines of the cluster in a distributed File system (DFS) and is usually represented as (key,value) tuples. It uses parallel algorithm which partitions the data across the clusters and in general it supports a wide range of similarity measures.

Keywords

MapReduce, Parallel Algorithm, Similarity, Pairwise Comparison.

Full Text

     

T. Hariharan ¹, K. Mohana Sundaram ²

Affiliations
1 Department of EEE, Dhirajlal Gandhi College of Technology, Anna University, Salem – 636309, Tamil Nadu, IN
2 Department of EEE, Vel Tech Multi Tech, Anna University, Chennai - 600025, Tamil Nadu, IN

Abstract

Optimal Power Flow (OPF) is the important for operation and planning of the power system. The OPF results are used in operation to improve the quality and generation cost of the power system. In planning the results of OPF is used to forecast future electric load expansion based on past and current operating details. The generation cost in OPF may reduce by adjusting the control variables generated real power; voltage magnitude and transformer tap position. Flexible AC Transmission System (FACTS) devices are further helpful for reducing generating cost as well as to improve the quality of the power supply. This paper investigates the helpfulness of three types of FACTS devices namely, STATCOM, SSSC and UPFC. To find optimal values of these control variables including FACTS devices hybrid intelligent algorithm Real Coded Genetic Algorithm-Differential Evolution (RGA-DE) is used for solving OPF problem. RGA has good selection and crossover operator but lays less emphasis on mutation operator whereas DE has good mutation process. To validate the work with other published work, IEEE 30 bus system is to be considered for the simulation.

Keywords

DE, FACTS, Hybrid RGA-DE, NR, OPF, Power Flow, RGA

Full Text