Assistant Professor working at CHRIST University, Bangalore. Phone: +919986737856 Address: #112,WestVillage , Kodipalaya,Kengeri
less
Interests
Uploads
Papers by Merin Thomas
MuLSA - Multi Linguistic Sentimental Analyzer for Kannada and Malayalam using Deep Learning
2021 2nd International Conference on Communication, Computing and Industry 4.0 (C2I4), 2021
Natural language Processing has been always a topic of interest in artificial intelligence. Opini... more Natural language Processing has been always a topic of interest in artificial intelligence. Opinion mining or Sentiment Analysis is an important application of Natural language Processing. Sentiment Analysis of text is to extract the sentiments underlined in the text. In this paper, a multi-linguistic sentimental analyzer (MuLSA), is implemented, a model that would address Malayalam, Kannada and English text. This model explores two languages in three categories of the text, its original script, transliterated script, and the combination of both along with English. Deep Learning, Recurrent Neural Network with LSTM is used as the basis for this model. The model exhibits 82% of prediction accuracy.
International Journal for Scientific Research and Development, 2015
In this paper, we aim to provide protection and security solutions based on many aspects of large... more In this paper, we aim to provide protection and security solutions based on many aspects of large integrated system in cloud. Cloud computing is the emerging technology to minimize the lots of users burden by providing different types of service to users. Here we also implement our system using ns2 which also show how our application can work in real world scenario .First we list some of the sources where we can download ns3. Secondly we discuss how to build our application and implementing in a real world database. It is high-speed data recovery scheme with minimal loss probability and using a forward error correction scheme to handle bursty loss. The proposed approach is highly efficient in recovering the singleton losses almost immediately and from bursty data losses.
Domain based sentiment analysis in regional Language-Kannada using machine learning algorithm
2016 IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology (RTEICT), 2016
Sentiment analysis (SA) is one of the important fields of Machine learning Language which involve... more Sentiment analysis (SA) is one of the important fields of Machine learning Language which involves analysis using natural language processing. The main goal of sentiment analysis is to detect and analyze attitude, opinions or sentiments in the text. Sentiment analysis has reach edits popularity by extracting Knowledge from huge amount data present online. The Process of analysis includes selecting features and opinion which is a challenging task in languages other than English. There are very few research works done for determining sentiments in regional languages. This Paper aims on domain based sentiment analysis in Regional language specific to movie susing machine learning algorithm for classification and provide a comparison between analysis using direct Kannada dataset and machine translated English language.
Indonesian Journal of Electrical Engineering and Computer Science, 2020
World has become very small due to software internationationalism. Applications of machine transl... more World has become very small due to software internationationalism. Applications of machine translations are increasing day by day. Using multiple languages in the social media text is an developing trend. .Availability of fonts in the native language enhanced the usage of native text in internet communications. Usage of transliterations of language has become quite common. In Indian scenario current generations are familiar to talk in native language but not to read and write in the native language, hence they started using English representation of native language in textual messages. This paper describes the identification of the transliterated text in cross lingual environment .In this paper a Neural network model identifies the prominent language in the text and hence the same can be used to identify the meaning of the text in the concerned language. The model is based upon Recurrent Neural Networks that found to be the most efficient in machine translations. Languag...
Sentiment analysis has been an important topic of discussion from two decades since Lee published... more Sentiment analysis has been an important topic of discussion from two decades since Lee published his first paper on the sentimental analysis in 2002. Apart from the sentimental analysis in English, it has spread its wing to other natural languages whose significance is very important in a multi linguistic country like India. The traditional approaches in machine learning have paved better accuracy for the Analysis. Deep Learning approaches have gained its momentum in recent years in sentimental analysis. Deep learning mimics the human learning so expectations are to meet higher levels of accuracy. In this paper we have implemented sentimental analysis of tweets in South Indian language Malayalam. The model used is Recurrent Neural Networks Long Short-Term Memory, a deep learning technique to predict the sentiments analysis. Achieved accuracy was found increasing with quality and depth of the datasets.
In the recent years Sentiment analysis (SA) has gained momentum by the increase of social network... more In the recent years Sentiment analysis (SA) has gained momentum by the increase of social networking sites. Sentiment analysis has been an important topic for data mining, social media for classifying reviews and thereby rating the entities such as products, movies etc. This paper represents a comparative study of sentiment classification of lexicon based approach and naive bayes classifier of machine learning in sentiment analysis.
In the recent years Sentiment analysis (SA) has gained momentum by the increase of social network... more In the recent years Sentiment analysis (SA) has gained momentum by the increase of social networking sites. Sentiment analysis has been an important topic for data mining, social media for classifying reviews and thereby rating the entities such as products, movies etc. This paper represents a comparative study of sentiment classification of lexicon based approach and naive bayes classifier of machine learning in sentiment analysis.
With the development of Internet technologies, there is an enormous amount of info rmation that i... more With the development of Internet technologies, there is an enormous amount of info rmation that is getting accumu lated in the World Wide Web, such as reviews, blogs, tweets, posts etc. In recent years sentimental analysis has gained mo mentum due to the increase in the size of information. Sentimental Analysis has been important in reviewing products, movies etc. Firstly, we use web scraping to scrape informat ion about movie reviews fro m a website. Approach is to determine the sentiment polarity of Kannada movie reviews. This paper introduces a hybrid method including both Lexicon based and Machine Learning.
Sentiment analysis has been an important topic of discussion from two decades since Lee published... more Sentiment analysis has been an important topic of discussion from two decades since Lee published his first paper on the sentimental analysis in 2002. Apart from the sentimental analysis in English, it has spread its wing to other natural languages whose significance is very important in a multi linguistic country like India. The traditional approaches in machine learning have paved better accuracy for the Analysis. Deep Learning approaches have gained its momentum in recent years in sentimental analysis. Deep learning mimics the human learning so expectations are to meet higher levels of accuracy. In this paper we have implemented sentimental analysis of tweets in South Indian language Malayalam. The model used is Recurrent Neural Networks Long Short-Term Memory, a deep learning technique to predict the sentiments analysis. Achieved accuracy was found increasing with quality and depth of the datasets.
World has become very small due to software internationationalism. Applications of machine transl... more World has become very small due to software internationationalism. Applications of machine translations are increasing day by day. Using multiple languages in the social media text is a developing trend. Availability of fonts in the native language enhanced the usage of native text in internet communications. Usage of transliterations of language has become quite common. In Indian scenario current generations are familiar to talk in native language but not to read and write in the native language, hence they started using English representation of native language in textual messages. This paper describes the identification of the transliterated text in cross lingual environment. In this paper a Neural network model identifies the prominent language in the text and hence the same can be used to identify the meaning of the text in the concerned language. The model is based upon Recurrent Neural Networks that found to be the most efficient in machine translations. Language identification can serve as a base for many applications in multi linguistic environment. Currently the South Indian Languages Malayalam, Tamil are identified from given text. An algorithmic approach of Stop words-based model is depicted in this paper. Model can be also enhanced to address all the Indian Languages that are in use.
Sentiment analysis (SA) is one of the important fieldsof Machine learning Language which involves... more Sentiment analysis (SA) is one of the important fieldsof Machine learning Language which involves analysis using natural language processing.The main goal of sentiment analysis is to detect and analyze attitude, opinions or sentiments in the text. Sentiment analysis has reachedits popularity by extracting Knowledge from huge amount data present online. The Process of analysis includes selecting features and opinionwhich is a challenging task in languages other than English. There are very few research works done for determining sentiments in regional languages. ThisPaperaims on domain based sentiment analysis in Regional language specific to moviesusing machine learning algorithm for classification and provide a comparison between analysis using direct Kannada dataset and machine translated English language.
Biometric systems for today's high security applications must meet stringent performance requirem... more Biometric systems for today's high security applications must meet stringent performance requirements. The fusion of multiple biometrics help to minimize the system error rates. In this paper, we are trying to propose a new model for the automated election in India using multimodal biometrics.
Cities and giant metropolis are a wonder to the average person. It takes several man hours to pla... more Cities and giant metropolis are a wonder to the average person. It takes several man hours to plan the urbanisation of an area and multiple amendments to make it fit the current as well as future demographic and political scenarios. It involves allotting areas for residential living, parks, public buildings (like post offices, hospitals, police stations etc.) and commercial areas. It also includes road networks to connect all the aforementioned places in a way that is efficient and convenient. This takes designers and planners weeks to do the right calculations, model the structures and assign them in an ideal way. Procedural generation is a method that helps to create content algorithmically and in large amounts. This paper aims to create an efficient workflow that attempts to lay out a terrain, blocks and other components to generate a full city. It consists of three work modules namely, designing the buildings algorithmically, coding lifelike textures and craft a competent road network. The modules are then integrated to display a city that can be altered by editing a seed value to generate and display another blueprint. This way, the paper aims to achieve multiple city visualizations for comparison and selection. The techniques used in this paper such as Perlin noise, fractals and L-systems help to realise the aim.
Uploads
Papers by Merin Thomas