Shruti Ranjan Satapathy

Shruti Ranjan Satapathy

Bengaluru, Karnataka, India
2K followers 500 connections

Join to follow

About

I am primarily interested in product-engineering architecture and operations, with a…

Activity

Join now to see all activity

Experience

  • Porter Graphic

    Porter

    Bengaluru, Karnataka, India

  • -

    Bengaluru, Karnataka, India

  • -

    Bengaluru Area, India

  • -

    Bengaluru, Karnataka

  • -

    Redwood City, California

  • -

    Bengaluru Area, India

  • -

    Bengaluru Area, India

Education

  • Indian Institute of Technology, Kanpur Graphic

    Indian Institute of Technology, Kanpur

    -

    Did my thesis on Word Sense Disambiguation.
    - The thesis proposes a supervised and a graph­ based solution to all-­words word sense disambiguation.
    - The supervised approach makes use of a mix of syntactic and semantic features, using SVMs for training.
    - The graph­ based approach makes use of contextual Lexical Knowledge Base (LKB) graph projections, applying PageRank’s powerful notion of graph centrality for ranking word senses.

  • -

    Activities and Societies: Music Club, Rubik's Cube Hobby Group, Junior Editor of ENCRE (student magazine IIT Kanpur)

Publications

Courses

  • Algorithmic Information Theory

    CS687

  • Algorithms-II

    CS345

  • Artificial Intelligence Programming

    CS365

  • Compiler Design

    CS335

  • Computer Networks

    CS425

  • Data Mining

    CS685

  • Data Structures and Algorithms

    ESO211

  • Discrete Mathematics

    CS201

  • Introduction to Computer Organisation

    CS220

  • Introduction to Mathematical Logic

    CS202

  • Machine Learning: Tools and Techniques

    CS698O

  • Multi-agent Systems: Games, Algorithms, Evolution

    CS785

  • Operating Systems

    CS330

  • Principles of Database Systems

    CS315

  • Principles of Programming Languages

    CS350

  • Probability and Statistics

    ESO209

  • Programming Tools and Techniques

    CS355

  • Randomized Algorithms

    CS648

  • Theory of Computation

    CS340

Projects

  • Semantic Textual Similarity

    -

    - The objective of the project was to design a generic semantic text similarity metric.
    - We modeled the problem into the “Weighted Bipartite Graph Matching” paradigm, making use of distributional word similarity measures at the core, learnt from large generic text corpora like Wikipedia.
    - The work resulted in a publication at the Semantic Evaluation 2012 Workshop, held in association with *SEM 2012­ First Joint Conference on Lexical and Computational Semantics, Montreal, Canada.

    Other creators

  • Emotion Detection in Text

    -

    - The project aimed at exploring the text based emotion prediction problem using supervised learning techniques.
    - The goal was to perform both coarse-­grained (emotion polarity) and fine-­grained (identifying from a pre­-defined emotion set) classification of English sentences.
    - We applied the Multinomial Naive Bayes algorithm as the learning scheme, which made use of n­-gram emotion word markers, extracted using an automatic feature selection algorithm.

    Other creators

  • Using Higher Order Dependencies for Sentiment Analysis of Text

    -

    - The project aimed at automatically identifying the sentiment / polarity of textual data.
    - In order to perform deep semantic reasoning, structural features are exploited from text, extracted from dependency parse trees, in addition to lexical features, and Support Vector Machines are used to perform classification.
    - A dependency subgraph mining and generalization algorithm was proposed for syntactic feature extraction.
    - The system showed significant improvement over n­-gram word…

    - The project aimed at automatically identifying the sentiment / polarity of textual data.
    - In order to perform deep semantic reasoning, structural features are exploited from text, extracted from dependency parse trees, in addition to lexical features, and Support Vector Machines are used to perform classification.
    - A dependency subgraph mining and generalization algorithm was proposed for syntactic feature extraction.
    - The system showed significant improvement over n­-gram word features, emphasizing the fact that the syntax of text is an important factor in disambiguating its sentiment.

  • Email Authorship Detection

    -

    - The aim of the project was to use computational stylistics for performing authorship identification of electronic mails.
    - We explored different linguistic style markers as features with a selection criterion imposed for automated feature pooling.
    - Finally, the features are used to classify anonymous emails using multi­-class variants of Support Vector Machines.

    Other creators

Honors & Awards

  • Academic Excellence Award

    Indian Institute of Technology Kanpur

    Was awarded the academic excellence award for the academic session 2009-2010.

  • National Camp for the Indian National Astronomy Olympiad 2006

    Indian National Astronomy Olympiad

    Was selected (top 30) to the national camp for the Indian National Astronomy Olympiad 2006.

Test Scores

  • All India Engineering Entrance Examination

    Score: All India Rank 250

    Got a rank of 250 out of ~ 1000000 applicants.

  • Indian Institute of Technology Joint Entrance Examination

    Score: All India Rank 179

    Got a rank of 179 out of ~ 500000 applicants.

Languages

  • English

    Full professional proficiency

  • Hindi

    Native or bilingual proficiency

  • Oriya

    Native or bilingual proficiency

  • Telugu

    Limited working proficiency

More activity by Shruti Ranjan

View Shruti Ranjan’s full profile

  • See who you know in common
  • Get introduced
  • Contact Shruti Ranjan directly
Join to view full profile

Other similar profiles

Explore more posts

Explore collaborative articles

We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.

Explore More

Add new skills with these courses

See all courses