Download CV

Research & CV

Mayesha Maliha Proma

Human-AI Interaction Researcher | Data Scientist | AI Product Manager | Founder

My research interests lie at the intersection of human-centered AI, decision-support systems, and the real-world evaluation of AI technologies. I am particularly interested in how people interpret, trust, and act on AI-generated guidance in consequential decision-making contexts.

NLP & Large Language ModelsTransformers & RAGRecommendation SystemsAI Agents & Knowledge GraphsAutomated Writing EvaluationHuman-AI Decision-MakingCausal Inference & Field ExperimentsEdTech & AI Decision Support
WritingPublicationsContact Me
Portrait of Mayesha Maliha Proma
Mayesha, in ink

pinned, not framed

Research Interests

AI for product decision-making, causal evaluation of user behavior, personalized recommendation systems, and NLP for stronger personal essays.

Explore research →

Current Focus

Designing AI interventions in student-facing products and testing whether they improve matching, trust, booking decisions, engagement, and continued use.

See current projects →

Education

BSc in Digital Business and Innovation with a concentration in Data Science and AI from Tokyo International University. Full-tuition scholarship. Preparing for thesis-based graduate research.

View education →

Selected Links

Google Scholar · LinkedIn · AbroadMates · Mentorship · Writing · Medium

See all links →

Research Interests

Human-centered AI, causal field evaluation, and the real-world impact of AI-enabled systems, with a focus on user behavior, trust, decision-making, and product outcomes.

Causal Field Evaluation of AI Systems

Studying whether AI models and agents achieve their intended goals when deployed in real products, which users benefit, and under what conditions they work best.

Causal InferenceField ExperimentsAI Evaluation

Human-AI Decisions and User Behavior

Examining how recommendations, explanations, personalization, and human support influence decisions, trust, adoption, conversion, engagement, and retention.

Human-AI InteractionDecision SupportUser Behavior

Language Models and Intelligent Systems

Developing and evaluating language-based systems for writing assessment, retrieval-augmented generation, knowledge-based applications, and personalized AI guidance.

NLPLLMsAI AgentsEvaluation

Ongoing Research

Ongoing Research

Causal Field Evaluation of AI Systems in Study-Abroad Platforms

Adapts causal field evaluation methods to AI models and agents deployed in real study-abroad products. The research examines whether these systems achieve their intended goals and how recommendations, explanations, personalization, and human support affect decisions, trust, adoption, conversion, engagement, and retention.

My role

Lead researcher

Current stage

Active research and experiment design

Methods

Randomized Field Experiments · A/B Testing · Causal Inference · Behavioral Analytics

Causal AI EvaluationField ExperimentsHuman-AI InteractionUser Behavior
Framework paper

Selected Publications & Manuscripts

View publication record →
Manuscript in Revision

Do Document Boundaries and Rhetorical Transitions Matter for Holistic Essay Scoring?

Tests whether discourse-derived document structure adds predictive value beyond essay text, surface features, and component counts on PERSUADE 2.0. Full-structure features improved held-out QWK from 0.836 to 0.849 and outperformed the baseline across all 15 prompts.

My role

Lead researcher and first author

Methods

PERSUADE 2.0 · TF-IDF + Ridge · DeBERTa · QWK · Ablation

Automated Essay ScoringNLPDocument Structure
GitHub
Published

Survey of Graph-Based Text Summarization for HRM Tasks

Reviews graph-based text summarization methods for human resource management tasks and synthesizes the main approaches, applications, and research directions.

NLPIR 2024 · Proceedings of the 8th International Conference on Natural Language Processing and Information Retrieval · pp. 380–387

My role

Co-author

Methods

Literature Review · Graph-Based Summarization · NLP · HRM

Text SummarizationGraph MethodsNLPHRM
Paper
Published

Augmenting Patent Summarization Using a Large Language Model with a Knowledge Graph

Combines LLaMA 3 with patent knowledge graphs and evaluates the generated summaries through ROUGE and human assessment. Human evaluators preferred the knowledge-graph summaries in most tested patent cases, although the baseline achieved higher ROUGE scores.

LKM@IJCAI 2024 · First International OpenKG Workshop: Large Knowledge-Enhanced Models · pp. 1–13

My role

Co-author

Methods

LLaMA 3 · Knowledge Graphs · Neo4j · ROUGE · Human Evaluation

Patent SummarizationLLMsKnowledge GraphsNLP
Paper

Products & Research Platforms

Alongside my research, I co-founded The Abroad Company and work on products that support students throughout the study-abroad journey. My work spans human mentorship, AI-supported application planning, product strategy, and user experience.

Parent Company

The Abroad Company

The Abroad Company builds products for students navigating study abroad and international mobility.

My role

Co-founder. I lead product direction, research, and the development of connected services across the student journey.

Focus areas

  • Product strategy
  • Research
  • International mobility
  • Growth and partnerships

AbroadMates

Study-Abroad Mentorship Platform

AbroadMates connects students with mentors who have already studied, applied, received scholarships, or lived in their target countries.

What we do

  • Find the right mentor

    Search by university, country, scholarship, field, and experience.

  • Book and meet

    Easy scheduling for 20 or 40-minute mentorship sessions.

  • Get real guidance

    Receive honest advice from someone who has already walked the path.

Research setting

Helps me study human guidance, mentor matching, trust, repeat use, retention, and hybrid human-AI support in real-world settings.

Learn more

ApplicationMate

Study-Abroad Application Copilot

ApplicationMate helps students organize and complete their applications through AI-powered guidance and structured workflows.

What we do

  • Discover and plan

    Receive personalized university and scholarship recommendations.

  • Organize everything

    Manage documents, tasks, deadlines, requirements, and reusable information in one place.

  • Write and improve

    Use AI-supported writing feedback and application guidance.

Research setting

Enables experiments on AI recommendations, agents, decision quality, task completion, adoption, engagement, and retention.

Learn more

These products create real-world environments for studying how human and AI support shape decisions and outcomes.

Writing & Research Notes

My writing includes beginner-friendly essays on artificial intelligence and technology, alongside an evolving research notebook where I document papers, questions, field experiment ideas, and reflections related to my current work.

Published Essays

LEARNING IN PUBLIC

Short Blogs on AI & Technology

These essays document how I learned technical ideas by explaining them in simple language and connecting them to tools, experiments, and products I was building. Together, they trace my journey from foundational data science concepts to neural networks, language models, and applied AI systems.

View all on Medium ↗

Research Notebook

AI & Education Reflections

An evolving research notebook containing paper notes, open questions, experimental ideas, and reflections from ongoing work with AI-enabled systems.

Open research notebook →

Experience & Education

My journey brings together human-centered AI, research, product building, teaching, and student leadership — from undergraduate research and founding a data science club to co-founding The Abroad Company.

Professional Experience

English Instructor

Maple English Club & Cafe, Maple Inc.

  • Teach English full-time to children and Japanese professionals.
  • Plan and deliver lessons for different ages, proficiency levels, and communication goals.
English TeachingCommunicationJapan

Co-founder and Product Lead

The Abroad Company

Co-founded The Abroad Company and lead the development of products supporting students throughout the study-abroad and international university application journey.

  • Lead product strategy, user experience, content, research direction, and early-stage growth.
  • Oversee the development of AbroadMates, a study-abroad mentorship platform.
Co-founderProduct LeadershipAI SystemsStudy Abroad

Education

BSc in Digital Business and Innovation

Tokyo International University

Completed a Bachelor of Science in Digital Business and Innovation with a concentration in Data Science and AI.

  • Recipient of a full-tuition scholarship.
  • Studied digital business, data science, artificial intelligence, machine learning, and product development.
  • Developed experience in NLP, data analysis, AI applications, research methods, and human-centered product design.
  • Worked as a Research Assistant for four semesters.
  • Served as a Teaching Assistant for two consecutive semesters.
  • Founded the TIU Data Science and Analytics Club.
  • Conducted research while simultaneously building technology products through The Abroad Company.
Full-Tuition ScholarshipData ScienceAIDigital Business

Skills & Methods

Research Methods

How I design and evaluate research.

Experimental Design · QWK Evaluation · Feature Ablation · Literature Synthesis · Causal Thinking

AI & NLP

Models and techniques I work with.

Transformers · RAG · Knowledge Graphs · AES · Embeddings

Product & Systems

Building things people actually use.

0→1 Product · User Research · EdTech UX · Cross-border Workflows

Tools

Day-to-day stack.

Python · PyTorch · MongoDB · Next.js · React · LaTeX · Git