Skip to content
View meghorikawa's full-sized avatar

Highlights

  • Pro

Block or report meghorikawa

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
meghorikawa/README.md

πŸ‘‹ Hi, I’m Megan (@meghorikawa)

  • 🌐 Multilingual: πŸ‡ΊπŸ‡Έ English (native) | πŸ‡―πŸ‡΅ Japanese (C1/C2) | πŸ‡©πŸ‡ͺ German (B1)
  • πŸŽ“ Computational Linguistics (M.A.) (expected 9/2025)
  • πŸ’‘ Research Areas of interest:
    • Linguistic Complexity
    • ICALL (Intelligent Computer-Assisted Language Learning)
    • Text Classification (sentiment, affect, learner error analysis)
    • Affordances & applications of Large Language Models
    • Educational Technology & Task-Based Learning

πŸ“– Master's Thesis

Modeling L2 Japanese Proficiency with Linguistic Complexity Measures and Criterial Features

  • Investigated how linguistic complexity measures and criterial features can model Japanese as a Second Language (L2) proficiency using the I-JAS corpus.
  • Applied an Explainable Boosting Machine (EBM) for interpretable classification

πŸ”­ Projects

  • KReLax: Multilingual Emotion Detection(SemEval-2025)
    Co-authored a published paper proposing an ensemble-based approach to multilingual emotion detection and addressing data imbalance. ACL Anthology (2025)
  • ARES (AI-Assisted Extension for Grammar Exercises)
    Enhancing grammar learning with AI-driven exercise generation aligned with pedagogical principles.
  • LLM Evaluation for Task-Based Language Learning Chatbots
    Analyzed the integration of LLMs into a rule-based chatbot to support smoother learner interactions within a language learning app.
  • Lingustic benchmark performance: Testing Effects of Reinforcement Learning
    Explored how reinforcement learning fine-tuning (RLHF) shifts LLM performance on syntax, semantics, and pragmatics benchmarks compared to base models.

Pinned Loading

  1. LangDev LangDev Public

    Writing Complexity measures on a L2 Japanese learner Corpus

    Jupyter Notebook

  2. ULLM ULLM Public

    Evaluating the Affordances of Large Language Models for Enhancing Rule-Based Chatbots for Task-Based Language Teaching

    Jupyter Notebook

  3. Flesch-Kincaid Flesch-Kincaid Public

    a project for the winter 24/25 linguistic complexity seminar which calculates Flesch-Kincaid readability score across a corpus

    Python

  4. AIEDproject AIEDproject Public

    Our final project for winter '24 seminar of Education system design with AI

    Java

  5. JFE JFE Public

    A Japanese text feature extractor

    Jupyter Notebook

  6. HJYnoDebug/KReLaX HJYnoDebug/KReLaX Public

    Jupyter Notebook 1