YI YI CHAN MYAE WIN SHEIN

Projects

- Listing Recommendation ( Neural collaborative Filtering )

- Listing Valuation ( Regression )

- Room Categorization ( Computer Vision )

- Watermark Detection and Removal ( Image Processing )

- Primary Listing Detection ( NLP )

• Designed, experimented, built, evaluated and motinored ML models and MLOps pipelines to solve business problems
• Mainly used Google Cloud Platform GCP services such as BigQuery, Kubernetes, Cloud Run, Cloud Build, GCS
• Developed custom machine learning algorithms for specific industry needs, resulting in improved performance and efficiency.
• Collaborated with cross-functional teams to integrate machine learning solutions into existing software systems.
• Friendly used the monitoring and management tools, such as Superset, Gradio, Argo, Grafana, Jira

• Building up the state-of-the-art open science multilingual generative language model

Projects

- Fraud Detection on Time Series Data

- Burmese Hate Speech Detection

- Arabic Hate Speech Detection

- Multilingual Harmful Speech Detection ( English, Arabic, Burmese, Tamil, Sinhala)

• Designing and building quality datasets
• Data wrangling and text mining on low-resourced ASIAN languages
• Exploratory Data Analysis (EDA) on structured/unstructured datasets such as text data, both nominal and ordinal data, time-series data
• Designed and built machine learning architectures, models, system pipelines and RESTful APIs to do Hate Speech Detection, Harmful Speech Detection and Fraud Detection with the performance scores from ~ 87 % to ~90%
• Worked with BitBucket, Google Colab, Azure ML Platforms
• Applied different Linear Machine Learning models (Logistic Regression, SVM, Naive Bayes, KNN) and Deep Learning models (CNN, LSTM, GRU, Transformer models such as BERT)

Projects

- Semantic Keyword Search Engine

- Job Recommender System

- PDF CV/Resume Parser

- Business Insight Dashboard

• Retrieved relevant information from both tubular data and text data from the job vacancies and job postings
• Finding the important information from text data of the candidate's resume/ CV
• Searching and Ranking the relevant job postings based on the information of the respective candidate's portfolio
• Hosted RESTful APIs on the AWS and integrate with the system
• Designed a visualization dashboard for business insights under the consultation from GSMA

Projects

- Data Mining on Location Data ( Time Series Data )

- Targeted Advertisement on Age and Gender ( Structured Data )

- Word Segmentation on Burmese ( Unstructured Data )

- Sentence Segmentation on Burmese ( Unstructured Data )

- POS tagging on Burmese ( Unstructured Data )

- Topic Modeling ( Unstructured Data )

- Text Summarization ( Unstructured Data )

• Did Data Cleaning, Exploratory Data Analysis (EDA), Feature Engineering for multiple projects
• Applied ML algorithms such as Linear Regression, Logistic Regression, XG Boost, SVM, etc
• Implemented unsupervised N-gram based Word Tokenization algorithm
• Created Topic Modeling, Sentence Summarization and Word Summarization on texts to retrieve user interests using Latent Dirichlet Allocation ( LDA )
• Produced Keyword Clouds by applying Graph-based approaches such as PageRank and Pointwise Mutual Information ( PMI )
• Worked with Google Cloud Platform (GCP) for integrating with the systems using REST APIs