Kongu Engineering College

About: Kongu Engineering College is a based out in . It is known for research contribution in the topics: Cluster analysis & Control theory. The organization has 2001 authors who have published 1978 publications receiving 16923 citations.

Reduction of Voltage Harmonics in Solar Photovoltaic fed Inverter of Single Phase Stand Alone Power System

Abstract: In this paper, an optimal harmonic stepped waveform (OHSW) method is proposed to reduce the voltage harmonics available at the output of solar photovoltaic (SPV) fed fifteen level cascaded multilevel inverter (CMLI) This technique is used to solve the harmonic elimination equations based on stepped waveform analysis in order to obtain the optimal switching angles The OHSW method considers the output voltage waveform as four equal symmetries in each half cycle and the magnitude of six numbers of harmonic orders is reduced Simulations are carried out in Matlab/Simulink and a 3 kWp solar plant is implemented in hardware to show the effectiveness of the proposed system

A survey on the applications of Deep Neural Networks

Abstract: The real-world applications need someone who will process the data with precision and take respective actions and it has to be out there without any human intervention so basically the machines need to think and act, this is where the new era of Machine Learning, Deep Learning comes into place. Deep Learning gives the machines a way to think when trained from previous data and encounters. This paper will provide the application of Deep Learning in various fields.

A Hash based Mining Algorithm for Maximal Frequent Item Sets using Linear Probing

Abstract: . Data mining is having a vital role in many of the applications like market-basket analysis, in bio-technology field etc. In data mining, frequent itemsets plays an important role which is used to identify the correlations among the fields of database. In this paper, we propose an algorithm, HBMFI-LP which hashing technology to store the database in vertical data format. To avoid hash collisions, linear probing technique is utilized. The proposed algorithm generates the exact set of maximal frequent itemsets directly by removing all non- maximal itemsets. The proposed algorithm is compared with the recently developed MAFIA algorithm and is shown that the HBMFI-LP outperforms in the order of two to three. Key words : Mining-Frequent Item Sets-Hashing-Linear Probing-MAFIA etc (Received July 26, 2008 / Accepted January 19, 2009) 1. Introduction Frequent itemset mining has wide applications. The research in this field is started many years before but still emerging. This is a part of many data mining techniques like association rule mining, classification, clustering, web mining and correlations. The same technique is applicable to generate frequent sequences also. In general, frequent patterns like tree structures, graphs can be generated using the same principle. There are many applications where the frequent itemset mining is applicable. In short, they can be listed as market-basket analysis, bioinformatics, networks and most in many analyses. Agarwal et. al [4] is the first person to state this problem. Later many algorithms were introduced to generate frequent itemsets. Let I = { I

Modified closed loop model for service performance management

Abstract: Purpose – The purpose of this paper is to develop an integrated closed loop performance management model for service industries.Design/methodology/approach – The service performance of any organization is measured by considering qualitative and quantitative dimensions. The qualitative dimension includes the service quality factors. In order to measure the service quality precisely, fuzzy analytical hierarchy process (FAHP) has been employed in this paper. The data pertaining to both qualitative and quantitative dimensions are combined using extended Brown‐Gibson (EBG) model to measure the service performance. As an improvement process, fuzzy quality function deployment (FQFD) has been employed to redesign the existing services. A case study from automobile repair shops illustrates the usability of the model.Findings – The developed model quantifies service performance and ensures the improvement of the service process. The proposed model takes into account the uncertainty that occurs while capturing the s...

Sentiment Analysis using Feature Extraction and Dictionary-Based Approaches

Abstract: Natural Language Processing (NLP) is a study of computational treatment of human language in order to make it understandable for computers. It is used in the research fields like artificial intelligence, information engineering, statistics, sentiment analysis and linguistics. Sentiment analysis plays significant role in NLP which performs the series of operations computationally to identify and categorize the sentiment conveyed in segment of words. The proposed approach is to detect the polarity of words from twitter using feature extraction and dictionary-based methods. Method are compared with feature engineering techniques like CountVectorizer, TF-IDF, Word2Vec by using machine learning classifier and also scoring of each word is done by SentiWordNet, VADER dictionaries. From the results, it clarifies that feature extraction outperforms the dictionary-based method. The proposed model gives better accuracy to detect the polarity.