Posts by Collection

portfolio

Portfolio item number 1

Published: May 25, 2025

Short description of portfolio item number 1

Portfolio item number 2

Published: May 25, 2025

Short description of portfolio item number 2

projects

Socket programming

Published: November 25, 2020

Code to download files from server to client via TCP protocol (using sockets) with progress bar.

Linux C- shell - Operating Systems

Published: November 30, 2020

Developed a fully functional Linux shell in C that supports redirection, piping, environmental variables, and background and foreground process execution.

Brick-breaker-Game

Published: July 22, 2021

Built a Terminal based break breaker game using OOPS concepts.

Software-programming-for-performance

Published: July 22, 2021

Optimizing matrix multiplication using software optimization techniques.

• Led a team of five members in developing an automated IoT-based water level monitoring system using the HC-SR04 Ultrasonic distance sensor. • Designed and implemented a dynamic dashboard using MySQL, Django, React, and OM2M server.

Digital Image Processing - Python, OpenCV

Published: November 30, 2021

Implemented Spatial and Fourier domain image filtering, morphological operations, image segmentation, and image compression using Huffman Coding and edge Detection methods. Coded a fingerprint enhancement and feature extraction system.

SimpleRA - Data Systems

Published: November 30, 2022

Developed an RDBMS supporting database operations such as load, print, store, and transpose for integer-only tables, efficiently handling large-sized matrices.

Pictionary AI Agents

Published: June 20, 2023

• AI Agents: Developed AI agents for the multi-modal game Pictionary using generative AI models: Stable Diffusion and Dall-E mini to generate images for drawing agent. • Gameplay Analysis: Conducted player behavior analysis using game telemetry data to develop human-like AI agents. Derived feature sets, collected data, designed surveys, performed exploratory data analysis and statistical testing, resulting in a publication.

HappyDB Emotion Detection - NLP

Published: November 30, 2024

Performed sentiment analysis on the HappyDB dataset for emotion classification achieving 89.6% accuracy, comparing the performance of various models including Gradient-Boosting methods, BiLSTM, Naive Bayes, Random Forest and Regression.

Recommender Systems and Web Mining - Python

Published: December 01, 2024

Developed and optimized recommender systems for Goodreads book reviews and Ulta product ratings, and ranked in the top 35% in leaderboard. Enhanced prediction accuracy using techniques like Gradient Boosting, SVD, Jaccard similarity, and bias-based models.

Navigation and Mapping for Mobile Robots - ROS2, Python

Published: December 10, 2024

Developed and implemented algorithms in ROS2, Python for the Qualcomm RB5 robot navigation, including SLAM, path planning, landmark-based localization, and full-coverage mapping.

Follow-me Robot:

Published: January 10, 2025

Developed a Follow-Me robot using the Qualcomm RB3 board and wave rover robot, integrating pose estimation and SLAM for autonomous tracking and navigation. We presented our work at the Qualcomm University Platforms Symposium 2025.

Key Features:

Qualcomm RB3 Robotics Platform: Leveraged for edge-AI processing.
Pose Estimation: To identify and track the target.
ROS 2 (Humble): Middleware for modular system integration.

publications

“Draw Fast, Guess Slow”: Characterizing Interactions in Cooperative Partially Observable Settings with Online Pictionary as a Case Study

Published in Human-Computer Interaction – INTERACT 2023, 2023

Cooperative human-human communication becomes challenging when restrictions such as difference in communication modality and limited time are imposed. We use the popular cooperative social game Pictionary as an online multimodal test bed to explore the dynamics of human-human interactions in such settings. As a part of our study, we identify attributes of player interactions that characterize cooperative gameplay. We found stable and role-specific playing style components that are independent of game difficulty. In terms of gameplay and the larger context of cooperative partially observable communication, our results suggest that too much interaction or unbalanced interaction negatively impacts game success. Additionally, the playing style components discovered via our analysis align with select player personality types proposed in existing frameworks for multiplayer games.

Recommended citation: Kannan, K., Rajendran, A., Alluri, V., & Sarvadevabhatla, R. K. (2023, August). “Draw Fast, Guess Slow”: Characterizing Interactions in Cooperative Partially Observable Settings with Online Pictionary as a Case Study. In IFIP Conference on Human-Computer Interaction (pp. 283-303). Cham: Springer Nature Switzerland. https://link.springer.com/chapter/10.1007/978-3-031-42286-7_16

teaching

Presenting Text Stimuli For Production Experiments in PsychoPy Using the Builder

Handouts, Phonetics Lab, 2012

This is a very clear tutorial to creating a basic text-based experiment in PscyhoPy using the builder. Includes pictures of the interface and very simple, easy-to-follow directions. Designed for someone who has never used PsychoPy before.

Bayesian Statistics for Linguists

Slides, Phonetics Lab, 2013

This is a basic introduction to Bayesian Statistics for Linguists, which covers the fundamental differences between Bayesian and Frequentist statistics, a very shallow introduction to the Bayes theorem and some additional resources.

Linguistics Outreach

Posters & Activicties, Pacific Science Center, 2014

These materials were developed for Paws on Science; an annual event put on by the Pacific Science Center to help UW scientists connect with the public. They were designed with an elementary-school audience in mind.

Active Learning and Presenting Research

Handouts, TA/RA Conference on Teaching, Learning and Research, 2014

Handouts prepared for the annual TA/RA Conference on Teaching, Learning and Research at the University of Washington on incorporating active learning strategies in the classroom and presenting research. Geared towards graduate students in their first few years.

Anandhini