Ingroj Shrestha
Ingroj Shrestha
PhD Student
Department of Computer Science
The University of Iowa
Office: 317 MLH
Advisor: Padmini Srinivasan

Research Interests

Text Mining (Biomedical application psychiatry), Natural Language Processing, Debiasing, Model Explanation


Shrestha, Ingroj, and Padmini Srinivasan. "Comparing Deep Learning and Conventional Machine Learning Models for Predicting Mental Illness from History of Present Illness Notations." AMIA Annual Symposium Proceedings. Vol. 2021. American Medical Informatics Association, 2021.

Shrestha, Ingroj, and Shreeya Singh Dhakal. "Fine-grained part-of-speech tagging in Nepali text." Procedia Computer Science 189 (2021): 300-311.

Shrestha, Ingroj, and Jonathan Rusert. "NLP_UIOWA at SemEval-2020 Task 8: You’re Not the Only One Cursed with Knowledge-Multi Branch Model Memotion Analysis." Proceedings of the Fourteenth Workshop on Semantic Evaluation. 2020.

Shrestha, Ingroj, and Shreeya Singh Dhakal. "A new stemmer for Nepali language." 2016 2nd International Conference on Advances in Computing, Communication, & Automation (ICACCA)(Fall). IEEE, 2016.

Shrestha, Ingroj, Dhakal, Shreeya, Kadariya, Madan. "A Comparative Study of Stemming Algorithms for Nepali Language". Proceedings of 3rd International IT conference on ICT for Intelligent Computing. Kathmandu, Nepal, 2016.


Debiasing racial bias in hate speech detection

Selling Drugs with Online Behavioral Advertising: Targeting Consumers Emotions on YouTube

- Assess whether online behavioral advertisements for prescription drugs on YouTube utilize Google search-query-derived data that identifies the emotional states of individual consumers, quietly matching consumers with ads calibrated to chime with their private thoughts, eroding consumer self-determination and implicating the autonomy interest in the process.
- Built a pipeline to collect YouTube video ads (using Selenium), automate Google account creation (using PVACreator and, generate fake personas (positive: users searching heart disease and happy emotion bearing terms and negative: users searching heart disease and negative emotion bearing terms and neutral: searching only heart disease) and automate Google search for the created account
- Analyze the emotion in the ads received by positive and negative group

Sentiment Analysis of Nepali Texts

- Extracted Nepali news and tweets and developed a sentiment corpus for Nepali language. Implemented Naive Bayes algorithm to perform sentiment analysis of Nepali texts. Visualized the results using Sinatra Google charts and Google map.

Teaching Experience

Fall 2018

CS 2420 Database for Informatics (Prof.: Raman Aravamudhan)

Spring 2019

CS 4400:0001 Database System (Prof.: Palani Andiappan)
CS 4400:0002 Database System (Prof.: Raman Aravamudhan)

Fall 2019

CS 2420 Database for Informatics (Prof.: Raman Aravamudhan)
CS 4400 Database System (Prof.: Palani Andiappan)

Spring 2020

CS 3330:0001 Algorithms (Prof.: Denise Szecsei)

Summer 2020

CS 2210 Discrete Structures (Prof.: Anthony Delaubenfels)
CS 3330:0001 (Prof.: Denise Szecsei)

Falll 2020

CS 5110:0001 Introduction to Informatics (Prof.: Padmini Srinivasan)
CS 4400 Database System (Prof.: Palani Andiappan)

Spring 2021

CS 3330:0001 Algorithms (Prof.: Denise Szecsei)

Open Source Contribution

Nepali Support for FTS in PostgreSQL (Snowball Stemmer) [code] [slide]