Ayush kumar mishra

Indian Institute of Technology, Kanpur

Master of Science - MS(R), Computer science and engineering

January 1, 2020 – January 1, 2023

Oracle

Data scientist 2

August 1, 2024 – Present

Bengaluru, Karnataka, India · Hybrid

Carelon Global Solutions India

Data Scientist

August 1, 2023 – August 1, 2024

Bengaluru, Karnataka, India · Hybrid

Amazon

Applied Scientist

February 1, 2023 – July 1, 2023

Bengaluru, Karnataka, India · Hybrid

Multilingual Named Entity Recognition and Part of Speech Tag

January 1, 2022 – Present

• Design and development of Multilingual NER and POS systems for multiple Indian languages, mainly Hindi, Bengali, Telugu, Malayalam, and Marathi, with an emphasis on detecting semantically ambiguous and complex things in brief and low-context scenarios is the objective of this project. • Experimented with different models, namely HMM, Bi-LSTM CRF, Bi-LSTM Attention CRF, and mBERT-CRF • mBERT-CRF has done exceptionally well and provided accuracy comparable to or exceeding state-of-the-art.

Quora Question Pairs Similarity

October 1, 2021 – November 1, 2021

• Classifying whether a question pair is duplicated is the objective of this research. The challenge is adding extra features to the dataset for higher accuracy. • Various Machine Learning Models, such as Random Forest Classifier and XGBoost, are used to classify tasks. • At each step, additional features are added to the provided dataset, and the variation in accuracy for both models is analyzed. Obtaining a maximum accuracy of 78.5%

Covid 19 Data Analysis and Forecasting month’s prediction

July 1, 2021 – September 1, 2021

• Analysis of COVID-19 spread across the globe: The observed virus spread gained momentum in the first week of February 2020 and increased exponentially afterward. How did the virus propagate from Mainland China to the Rest of the world? Analyzing the pattern of deaths, Nation’s Competency to tackle corona via Analyzing Patient Recovery progression. • Analysis of COVID-19 in INDIA: Visualizing State-wise spread of Coronavirus and analyzing count of covid-19 cases by date. Investigating Cases with and without Travel history showed that 23% of total patients tested positive on recent travel. • Forecasting Month’s prediction in INDIA using SVM(R), Linear Regression(Polynomial), and Bayesian Ridge Regression. • Used Mean Absolute Error(MAE) and Mean Squared Error(MSE) for analysis, and polynomial LR gives the best result.

Aitihasikam: A Knowledge Graph Dataset for Querying over Ramayana

June 1, 2021 – December 1, 2021

• This work involved manually annotating the Valmiki Ramayana corpus and creating a knowledge graph using the annotations. • Using the knowledge graph, build a Question Answering System for giving relevant answers. • Worked on Named Entity Extraction, Entity Relation Extraction, and Question Pair similarity.

Neural Machine Translation System

March 1, 2021 – May 1, 2021

• Designed Neural Network models for Hindi to English Machine Translation and trained on 100K+ Hindi Sentences. • Implemented RNNs, LSTMs, Bi-GRUs, and Transformers NMT models with an Attention mechanism. • Achieve BLEU score and METEOR score of 1.309 and 0.389, respectively on a test set of 24K+ Hindi Sentences

Neural Network and Deep learning

Coursera

June 25, 2026 – Present

Machine Learning

Stanford University

June 25, 2026 – Present