Qualification: 
M.Tech, B-Tech
thara@am.amrita.edu

Thara S. currently serves as Assistant Professor at the Department of Computer Science and Engineering, School of Engineering, Amritapuri Campus. She pursued her M. Tech. degree in Computational Engineering and Networking. Prior to this, she worked as Project Engineer at Amrita Center for Cyber Security and Networks, Amrita Vishwa Vidyapeetham, Amritapuri Campus. She has 4 years and 8 months of academic experience.

Awards/Achievements

  • "Best Poster Paper Award" in the Seventh International Conference on Advances in Computing, Communications and Informatics (ICACCI, 2018)

Publications

Publication Type: Journal Article

Year of Conference Title

2018

S. Thara and Krishna, A., “Aspect Sentiment Identification using random Fourier features”, International Journal of Intelligent Systems and Applications (IJISA), vol. 10, pp. 32-39, 2018.[Abstract]


The objective of the paper was to show the effectiveness of using random Fourier features in detection of sentiment polarities. The method presented in this paper proves that detection of aspect based polarities can be improved by selective choice of relevant features and mapping them to lower dimensions. In this study, random Fourier features were prepared corresponding to the polarity data. A regularized least square strategy was adopted to fit a model and perform the task of polarity detection Experiments were performed with 10 cross-validations. The proposed method with random Fourier features yielded 90% accuracy over conventional classifiers. Precision, Recall, and F-measure were deployed in our empirical evaluations.

More »»

Publication Type: Conference Paper

Year of Conference Title

2018

S. Thara and Poornachandran, P., “Code-Mixing: A Brief Survey”, in 2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI), Bangalore, India, 2018.[Abstract]


Indians and many other non-English speakers across the world, prefer not to use single code in their messaging texts on social media platforms. They make use of transliteration and randomly merged English words using code-mixing, two or more languages to show their linguistic proficiency (English-Spanish, Arabic-English, etc.). Code-mixing (CM) is a dynamically progressive area of research in the domain of text mining. Present time communications in social media, blogs, reviews are abuzz with creative, crafty code-mixed messages. This paper highlights a comprehensive study of CM in the diverse fields of Natural Language Processing (NLP) including language identification, Part-of-Speech (POS) tagging, Named Entity Recognition (NER), Polarity Identification, Question Answering. CM has also been sought after in studies involving Machine Translation, Dialect identification, Speech technologies etc. Most of the applications of code mixing are scrutinized and presented briefly in this survey. This study purports to articulate tends and, techniques pursued in languages used and also unique evaluation measures to give accuracy.

More »»

2018

L. Sravani, Reddy, A. S., and Thara, S., “A Comparison Study of Word Embedding for Detecting Named Entities of Code-Mixed Data in Indian Language”, in 2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI), Bangalore, India, 2018.[Abstract]


Communication has increased many-fold in the internet era, making social media a lively platform for the exchange of information. Most people use multiple or mixed languages in their conversations as they share contemporaneous information. Code Mixing is a technique which mixes two or more languages within a dialogue. The extraction of relevant and meaningful information from mixed set of languages poses a tedious exercise. The objective of the paper is to perform named entity recognition (NER), one of the challenging task in the domain of natural language processing. The method proposed herein explores a novel exhaustive comparison study, heretofore un-addressed among four word embedding approaches like Continuous Bag of Words model (CBOW), Skip gram model, Term Frequency and Inverse Document Frequency (TF-IDF) and Global Vectors for Word Representation (GloVe). These word vector representing schemes decipher the meaning of words in different dimensions, such as in code mixed language pair English-Hindi. These word vectors or feature vectors, computed from co-occurrences, yielded good cross-validation scores when compared with six conventional machine learning algorithms. The study reveals Tf-IDF is the best word embedding model yielding the highest accuracy for the small dataset. Precision, Recall, and F-measure were used as evaluation measures.

More »»

2018

I. Chaitanya, Madapakula, I., Gupta, S. K., and Thara, S., “Word Level Language Identification in Code-Mixed Data using Word Embedding Methods for Indian Languages”, in 2018 International Conference on Advances in Computing, Communications and Informatics (ICACCI), Bangalore, India, 2018.[Abstract]


In recent years, social media networking has grown to be a marvel of technology in our way of life. Facebook operates the world's leading web-based social networking system with over 2.19 billion clients(as of the first quarter of 2018). As its popularity increased, more individuals from all age demographics, have been accessing this growing phenomenon. Resultant usage of code-mixed data has become an all too common practice in the context of social media. The aim of our project was to identify different languages in the processing of code-mixed data. A comparison of different word embedding methods like Continuous Bag of Words (CBOW) and Skip-Gram models was used to generate feature vectors. These vectors are given as input to the machine learning algorithms like Support Vector Machine, Logistic Regression, K-Nearest Neighbors, Gauss Naive Bayes, Adaboost, and Random Forest which yielded in good cross-validation scores. The paper also reveals that Precision, Recall, F-Score, Micro and Macro averaging were used as evaluation measures.

More »»

2017

S. Thara and Sidharth, S., “Aspect based sentiment classication: Svd features”, in 2017 International Conference on Advances in Computing, Communications and Informatics (ICACCI), Udupi, India, 2017.[Abstract]


Research on sentiment analysis and classification is a hot research topic as it have application in several disciplines and domains. In this paper, the work is focused on classification of laptop and restaurant data set towards three different polarity categories such as positive, negative, neutral. Current work used Singular Value Decomposition(SVD) based feature for sentiment prediction as it can capture the latent relation among the data. The paper presents a comparison on classification performed using SVM via linear, polynomial and rbf kernel, naive bayes, simple logistics, random forest. Precision, recall, f1 score, accuracy are used as evaluation measure. During the evaluation it is found that the SVM with rbf and polynomial gave better classification result.

More »»

2014

G. Gressel, K., S., A, A., Thara, S., P., H., and Poornachandran, P., “Ensemble learning approach for author profiling”, in Proceedings of CLEF 2014, 2014.[Abstract]


With the evolution of internet, author profiling has become a topic of great interest in the field of forensics, security, marketing, plagiarism detection etc. However the task of identifying the characteristics of the author just based on a text document has its own limitations and challenges. This paper reports on the design, techniques and learning models we adopted for the PAN-2014 Author Profiling challenge. To identify the age and gender of an author from a document we employed ensemble learning approach by training a Random Forest classifier with the training data provided by PAN organizers for English language only. Our work indicate that readability metrics, function words and structural features play a vital role in identifying the age and gender of an author.

More »»

2013

B. G. Gowri, Hariharan, V., Thara, S., Kumar, S. S., and K. P. Soman, “2D Image data approximation using Savitzky Golay filter #x2014; Smoothing and differencing”, in Automation, Computing, Communication, Control and Compressed Sensing (iMac4s), 2013 International Multi-Conference on, 2013.[Abstract]


Smoothing and differencing is one of the major important and necessary step in the field of signal processing, image processing and also in the field on analytical chemistry. The search for an efficient image smoothing and edge detection method is a challenging task in image processing sector. Savitzky Golay Filters are one among the widely used filters for analytical chemistry. Even though they have exceptional features, they are rarely used in the field of image processing. The designed filter is applied for image smoothing and a mathematical model based on partial derivatives is proposed to extract the edges in images. The smoothing technique of SG filter offers an extremely simple aid in extracting the edge information. An approach using SG filter which can be applied in preserving edge information is one of the major tasks involved in the classification process in the domain of Optical Character Recognition. The paper is focused on designing the Savitzky Golay filter by using the concepts of linear algebra. The main objective of the paper is to portray a clear cut idea about Savitzky Golay filter and to study the design of Savitsky Golay filters based on the concepts of Linear Algebra.

More »»

Publication Type: Conference Proceedings

Year of Conference Title

2013

B. G. Gowri, Hariharan, V., Thara, S., Sowmya V., Kumar, S. S., and Dr. Soman K. P., “2D Image data approximation using Savitzky Golay filter - Smoothing and differencing”, 2013 IEEE International Multi Conference on Automation, Computing, Control, Communication and Compressed Sensing, iMac4s 2013. IEEE, Kochi, Kerala, pp. 365-371, 2013.[Abstract]


Smoothing and differencing is one of the major important and necessary step in the field of signal processing, image processing and also in the field on analytical chemistry. The search for an efficient image smoothing and edge detection method is a challenging task in image processing sector. Savitzky Golay Filters are one among the widely used filters for analytical chemistry. Even though they have exceptional features, they are rarely used in the field of image processing. The designed filter is applied for image smoothing and a mathematical model based on partial derivatives is proposed to extract the edges in images. The smoothing technique of SG filter offers an extremely simple aid in extracting the edge information. An approach using SG filter which can be applied in preserving edge information is one of the major tasks involved in the classification process in the domain of Optical Character Recognition. The paper is focused on designing the Savitzky Golay filter by using the concepts of linear algebra. The main objective of the paper is to portray a clear cut idea about Savitzky Golay filter and to study the design of Savitsky Golay filters based on the concepts of Linear Algebra. © 2013 IEEE.

More »»