Hi there, my name is Shinjini Ghosh!
I am a senior at Massachusetts Institute of Technology, double majoring in Computer Science & Engineering and Linguistics, with minors in Mathematics and Brain and Cognitive Sciences. I plan to continue an MEng at MIT EECS with an Artificial Intelligence concentration. I am interested in research in machine learning, especially healthcare-focused, as well as in both pure and computational linguistics.
I love to spend my free time working on other software development projects ranging from data modeling in finance, healthcare and natural language processing to educational & scientific development tools, helpful extensions and fun games.
-
Automated Detection of SepsisAchieved 94% precision and 78% accuracy in detecting sepsis from neutrophil movement in microfluidic images using CNNs, LSTMs & GRUs
-
Detecting the Risk of Sepsis in COVID-19 Affected PatientsDeveloped deep learning models to detect the risk of sepsis and of needing hospitalization in COVID-19 affected patients, especially for the younger population
-
End-to-end Actuation in Self-Driving VehiclesAnalysed the prediction confidence of deep learning models from probability density functions
-
Study of Syntax Acquisition for Third, Fourth and Further LanguagesResearch paper on Third and Fourth Language Acquisition of Relative Clause Structures under works; designed and implemented an elicited imitation experiment and performed extensive data analysis to investigate the primacy of the free relative clause in the acquisition of complex structures
-
Computational Modeling of Child Language AcquisitionTo develop and implement a computational model for how children acquire a language that is grounded in contemporary theories of human language syntax known as "minimalism"
Publications
-
Ghosh S., Amini A., Rus D., From Data-to-Decisions - Learning Representations for End-to-end Sepsis Detection, presented at NeurIPS WiML Workshop, Dec 2020, won Best Poster Award at STEMM CSAIL AI in Healthcare Summit, Oct 2020.Upcoming
-
Ghosh S., Flynn S., Third Language Acquisition of Relative Clause Structures in Hindi, accepted to Berkeley Annual Linguistics Symposium, April 2021 and MIT Undergraduate Research Journal (MURJ), December 2020.Upcoming
-
Cherukuri A., Ghosh S., Quinteno-Rivera F., Trends of Incidence and Survival of Multiple Myeloma in the Adolescent and Young Adult (AYA) Population in the US, 2020 Lymphoma, Leukemia & Myeloma Congress - An International Congress on Hematologic Malignancies. Published in the American Journal of Hematology, 2020.Multiple Myeloma (MM) is a plasma cell malignancy that is common in older adults. According to “Cancer Facts & Figures of 2020” by the American Cancer Society, about 32,270 new cases of Myeloma will be diagnosed in 2020. However, the incidence and survival of Multiple Myeloma in the adolescent and young adult (AYA) population, ages 15-39 years, has not been thoroughly examined. In this study, we assessed the incidence, sex and race-specific distribution of Myeloma in the AYA population, and compared the relative survival rates of AYA patients to patients 40 years or older.
-
Ghosh S., Predictive Model with Analysis of the Initial Spread of COVID-19 in India, International Journal of Medical Informatics, Volume 143, 2020. https://doi.org/10.1016/j.ijmedinf.2020.104262.Objective - The Coronavirus Disease 2019 (COVID-19) has currently ravaged through the world, resulting in over thirteen million confirmed cases and over five hundred thousand deaths, a complete change in daily life as we know it, worldwide lockdowns, travel restrictions, as well as heightened hygiene measures and physical distancing. Being able to analyse and predict the spread of this epidemic-causing disease is hence of utmost importance now, especially as it would help in the reasoning behind important decisions drastically affecting countries and their people, as well as in ensuring efficient resource and utility management. However, the needs of the people and specific conditions of the spread are varying widely from country to country. Hence, this article has two fold objectives - (i) conduct an in-depth statistical analysis of COVID-19 affected patients in India, (ii) propose a mathematical model for the prediction of spread of COVID-19 cases in India. Materials and Method - There has been limited research in modeling and predicting the spread of COVID-19 in India, owing both to the ongoing nature of the pandemic and limited availability of data. Currently famous SIR and non-SIR based Gauss-error-function and Monte Carlo simulation models do not perform well in the context of COVID-19 spread in India. We propose a 'change-factor' or 'rate-of-change' based mathematical model to predict the spread of the pandemic in India, with data drawn from hundreds of sources. Results - Average age of affected patients was found to be 38.54 years, with 66.76% males, and 33.24% females. Most patients were in the age range of 18 to 40 years. Optimal parameter values of the prediction model are identified (α = 1.35,N = 3 and T = 10) by extensive experiments. Over the entire course of time since the outbreak started in India, the model has been 90.36% accurate in predicting the total number of cases the next day, correctly predicting the range in 150 out of the 166 days looked at. Conclusion - The proposed system showed an accuracy of 90.36% for prediction since the first COVID-19 case in India, and 96.67% accuracy over the month of April. Predicted number of cases for the next day is found to be a function of the numbers over the last 3 days, but with an ‘increase’ factor influenced by the last 10 days. It is noticed that males are affected more than females. It is also noticed that in India, the number of people in each age bucket is steadily decreasing, with the largest number of adults infected being the youngest ones — a departure from the world trend. The model is self-correcting as it improves its predictions every day, by incorporating the previous day’s data into the trendline for the following days. This model can thus be used dynamically not only to predict the spread of COVID-19 in India, but also to check the effect of various government measures in a short span of time after they are implemented.
-
Ghosh S., Language Identification Based on the Variations in Intonation Using Multi-classifier Systems, Mining Intelligence and Knowledge Exploration. MIKE 2017. Lecture Notes in Computer Science, vol 10682. Springer.In this article we make use of the characteristics of tonal languages and machine learning methodologies to understand the patterns in them. Instead of analyzing the absolute pitch or frequency, we analyze how one tone transitions to another in speech. Features (namely, zero crossing count, short time energy, minimum formant frequency, maximum formant frequency) are extracted using the tonal transitions over segments of audio signals. We have developed a multi-classifier system using four classifiers, namely maximum likelihood estimate (MLE), minimum distance classifier (MDC), k-nearest neighbor (kNN) classifier and fuzzy k-NN classifier to automatically identify tonal languages from audio signals. Initially, each individual classifier is trained with existing known data represented by the extracted features. The trained classifier is then used for language identification. Results obtained from these classifiers are combined to generate the final output. Experiments are conducted using three different tonal languages, namely, Chinese, Thai and Vietnamese. The output reveals that the developed multi-classifier model is able to produce promising results. The extracted features produced better results in comparison to usually used frequency value (as a feature). Ensemble of classifiers is a better tool than using individual classifiers.
Class projects, personal research, pet projects and misc.
-
Language Identification Based on the Variations in Intonation Using Multi-classifier SystemsPeer-reviewed conference paper, published in 2017. Third Grand Award, Systems Software Category, Intel ISEF 2017, Los Angeles, USA. Grand Award, IRIS National Science Fair, 2016.
-
A Novel Machine Learning Approach for Determining the Confounding Factors for Cancer Identification --- An Integration of Neural Learning and Decision TreeFourth Grand Award, Systems Software Category, Intel ISEF 2018, Pittsburgh, USA. Grand Award, IRIS National Science Fair, 2017.
-
Coronavirus in IndiaOngoing project to track the spread of coronavirus in India, predict spread, and assess the effects of current combat measures. Research paper published at the International Journal of Medical Informatics.
-
Agreement Attraction Effects in Multilingual Language ModelsFinal project for 6.884, Fall 2020. We investigate subject-verb agreement and related attraction effects in both monolingual and codeswitched data across three languages (English, French, and Russian). We first establish that mBERT has a cross-linguistic notion of agreement and that agreement attraction effects occur. We then investigate how and speculate as to why the agreement attraction effects in codeswitched sentences compare and contrast with the monolingual ones and how the pattern of these effects varies with the language of the verb.
-
Human Cognition Based Word Segmentation ModelsFinal project for 9.660, Fall 2020. Segmentation of words from free speech or unsegmented text is an almost universally prevalent human skill. In this project, we build, implement and test three computational models of word segmentation based on human cognition - a probabilistic context-free grammar model, a probabilistic n-gram model with dynamic programming, and a statistical Viterbi algorithm based approach. We also investigate how they perform in comparison with human cognition experiments in similar conditions.
-
Grammatical Sketch of ManxFinal paper for 24.900, Fall 2019. Presented a grammatical sketch of Manx, an Indo-European, Celtic, reawakening language with no known L1 speakers and emerging L2 speakers. Developed corpus from fieldwork with Manx speakers and analyzed it to build a sketch of basic phonology, morphology and syntax of Manx.
-
ESP32-based Arcade Game Platform (team project)Final project for 6.08 Embedded Systems, Spring 2020. Built an interactive arcade game platform with Hangman, Pong and Pictionary, using only an ESP32, microphone, speakers and IMU.
-
NerdyNSS - A network storage system for ExtraNet (team project)Final project for 6.033 Computer Systems Engineering, Spring 2020. A network storage system with high reliability, scalability and performance by effectively utilizing limited storage space and bandwidth.
-
Using SMT Solver to Infer Phonological ProcessesFinal project for 24.918, Spring 2020.
-
Acquisition of ReduplicationFinal project for 24.904, Language Acquisition, Spring 2020 on Child Acquisition of Reduplication in Bangla
-
IPAtopeAn interactive and playful IPA (International Phonetic Alphabet) display developed as an educational tool. Features include filtering the phones by voicing, place and manner of articulation and properties, sorting the phones by symbol and sonority, different themes as well as attribute toggling.
-
HelpMe VS Code ExtensionTired of retyping the same errors into Google and StackOverflow? Can't remember the one amazing link you once used for an elegant one-liner? HelpMe is here to save the day! Bookmark useful questions and their corresponding links and look them up whenever you need them later, without needing to leave VS Code.
-
CoviDistDeveloping a system to centrally equitably manage essential supply allocation and distribution from donors and storehouses to those who need them the most. Includes algorithmic component for efficient and equitable assignment considering distribution and transportation limits as well as application for usability.
-
Agent-based Epidemic Simulation(Upcoming)
-
The Next GENErationOur team aims to refine existing processes for precision medicine and genomic/proteomic analysis through developing an app that uses encryption to protect patient's privacy, as well as create a network for both researchers and patients.
-
Visualizing Simple Math and Science by GreenfootUtilizing game development software to build math and science simulations for school students. Developed during 2B-KMUTT International Summer Research Programme, 2016.
Teaching/Lab Assistant
Fall 2021
Spring 2021
IAP 2021
Fall 2020
Spring 2020
Fall 2019
Spring 2019
IAP 2019
Fall 2018
ASE (Advanced Standing Examinations)
Current Positions
Member of the Problem Committee and Jury, International Linguistics Olympiad
Since Aug 2018
- Designing, reviewing & testing problems for the International Contest, alongside 27 other linguists from around the world
- Grading solutions & deciding awards at the International Olympiad
- Multilingual editing & translating for the final problem set
Member of the Problem Committee and Jury, Asia-Pacific Linguistics Olympiad
Since Sep 2019
- Designing, reviewing & testing problems for the International Contest, alongside 8 other linguists from around the world
- Multilingual editing & translating for the final problem set
Member of the Problem Committee, Panini Linguistics Olympiad
Since Aug 2016
Designing, reviewing & testing problems for the various rounds of the National Contest, which serves to select the students representing Team India internationally at the International Linguistics Olympiad
Member of the New York Academy of Sciences
Since Aug 2018
- Mentor for the 1000 Girls, 1000 Futures Program
- Solving design & innovation challenges and conducting research in areas with high real-world impact
- Professional Member of the Academy
- Member of the Junior Academy
Teacher, MIT Educational Studies Program (ESP)
Since Nov 2018
- Co-taught a class on 'Conlangs - Create Your Own Language' to over 100 middle- and high-school students at MIT ESP's weekend program Splash in Fall 2018 and 6-week program HSSP in Spring 2019; weekend program Splash in Fall 2019 and Fall 2020
- To co-teach a class on Introduction to Git and GitHub in Spark, Spring 2021
- Co-designed the materials and worksheets necessary alongside 2 other undergraduate students
Student Clubs
I'm involved with the following activities and student clubs at MIT.
Associate Advisor for EECS Department
Associate Advisors serve as a resource for both Faculty advisors and undergraduate advisees in conjunction with the Undergraduate Office to help students and faculty be aware of available resources, construct three-year plans, facilitate strong relationships between undergraduates and their Faculty Advisors, promote an inclusive environment in EECS, and offer perspective about what it means to be a Course 6 student at MIT. I have been active in this position since Fall 2020.
Women in EECS (WiEECS) --- Professional Development Committee Executive Member
WiEECS's mission is to build a community for women in EECS that supports, encourages, and empowers them to succceed. On the Professional Development team, my responsibilities include organizing networking events with academicians and people in the industry, coordinating with organizations to hold information and recruiting events, and helping organize technical workshops. I have been active in this position since Spring 2019.
Student Alumni Association (MIT SAA) --- President (Former VP, Programming) & MIT Ambassador
The Student Alumni Association (SAA) is a student group that partners with the MIT Alumni Association to provide students with leadership and volunteer experiences consistent with their service as exceptional ambassadors of the Institute. As President of the organization since Summer 2021, my responsibilities include recruitment of new members, overseeing all organization activities, as well as ensuring smooth functioning. As VP of Programming, my responsibilities include coordination and planning for various student-alumni events throughout the semester. I have been active in this position since Spring 2019.
MITxHarvard Women in AI --- Executive Member, Interview Series
MIT Tau Beta Pi Engineering Honor Society --- Eligibles Chair
Undergraduate Society for Women in Math (USWIM) --- Executive Member
Member
- Undergraduate Association COVID-19 Committee --- gauge and assess issues faced by undergraduates, shape solutions, coordinate with and advise the decision making of MIT administrators
- Student advisory group for MIT's new Schwarzman College of Computing --- coordinate with the administration team and Dean of SCC to help shape the college's mission, principles, and workings
- Student advisory group for MIT EECS Department --- coordinate with EECS Department to help shape the EECS student experiences
- Society for Women Engineers (SWE) --- hold technical workshops for peers
- Staff Photographer and Meteorologist, The Tech, MIT's Campus Newspaper --- cover campus life weekly
- Maseeh Hall Committees
Past Positions
Teacher, Liceo Enrico Fermi, Livorno, Italy
January 2020
Teacher of Mathematics (with applications in Physics) and Computer Science for Junior Year, Senior Year Regular & Senior Year Advanced students for 3 sections each, and Junior Year Advanced & Senior Year Classico students for 1 section each -- developing materials, teaching, setting exams, assigning homework and helping students
Data Science & Engineering Intern, QuantCo, Berlin, Germany
Summer 2019
- Handled raw data to develop predictive models in the finance in healthcare sector
- Developed a web integration system
Student Researcher, King Mongkut's University of Technology, Thonburi, Bangkok, Thailand
Mar 2016 - Apr 2016
Visualizing Simple Math and Science by Greenfoot during 2B-KMUTT Camp 2016 (International Summer Research Programme), one of 20 international students selected
Get in touch!
Physical
I love receiving (and sending) letters and postcards! Please address to
Maseeh Hall 5017
Massachusetts Institute of Technology
305 Memorial Drive
Cambridge MA 02139
USA