Skip to content
View saeth40's full-sized avatar
5 followers · 1 following
Block or Report

Block or report saeth40

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
saeth40/README.md

📚Publication

1. Solving Cosine Similarity Underestimation between High Frequency Words by ℓ2 Norm Discounting

  • Proceedings of The 61st Annual Meeting of the Association for Computational Linguistics, Toronto, Canada, 2023
  • Supervised by Professor Danushka Bollegala, head of NLP and machine learning research group at the University of Liverpool
  • Co-author: Yi Zhou

🏆Competitions

1. Kaggle competition, Natural Language Processing with Disaster Tweets

  • Top 15% of the leaderboard globally by classifying potential disasters from extremely vague Twitter text dataset
  • Text preprocessing, BERT word and sentence embedding, fine-tuning BERT

💻Courswork/ projects

1. Masters in Data Science and AI, The University of Liverpool, UK

-> Database management and SQL, Applied AI (ML, DL), Math and Statistics for Data Science, Reinforcement learning and bioinspired optimization, Computational intelligence, Data mining and visualization, Object-oriented programing

2. Semi unsupervised NLP topic modeling for job classification

-> MongoDB, Gensim, NLTK, TF-IDF word embedding, Latent Dirichlet Allocation

3. End-to-end garbage sorting robot using object detection

-> Object detection, Transfer learning, Raspberry Pi, IoT server, OpenCV

4. AI for engineers, Chulalongkorn University, Thailand

-> Hill Climb, Genetic algorithm, K-mean, H-clustering, Regression, Decision tree, Naïve Bayes classifier, CNN transfer learning

🌱Self-learning skills

1. Completed PyTorch for NLP and deep learning

-> Transformer (BERT), Bi-directional LSTM, Time series analysis, Anomaly detection, Encoder-decoder (sequence to sequence), Text classification, Text summarization, Machine translation, Sentimental analysis, Named-entity recognition, Image augmentation, Image classification, Custom dataset preparation

2. AWS services

-> EC2 instance, S3 Bucket, Amazon SageMaker

3. Python for data analysis

-> Pandas, Numpy, Matplotlib, Seaborn, Data visualization

4. CSV table generation from web scraping and PDF table extraction

-> HTML, Beautiful Soup, Selenium

Pinned

  1. Semi-unsupervised-NLP-topic-modeling-Latent-Dirichlet-Allocation Semi-unsupervised-NLP-topic-modeling-Latent-Dirichlet-Allocation Public

    Query and preprocess more than 1 M data from MongoDB(NoSQL). Jobs classification and analysis to define skills demanded by the job market trend with unsupervised NLP and data visualization.

    Jupyter Notebook 1

  2. Garbage-Sorting-Robot-Using-Object-Detection Garbage-Sorting-Robot-Using-Object-Detection Public

    A user-friendly robot that can classify various types of garbage with high accuracy and work fast using object detection, Raspberry pi, Arduino, and IoT server to track system behaviors.

    Jupyter Notebook 1 3

  3. AWS-services AWS-services Public

    Jupyter Notebook

  4. Completed-PyTorch-for-NLP-and-deep-learning Completed-PyTorch-for-NLP-and-deep-learning Public

    Jupyter Notebook

  5. LivNLP/cosine-discounting LivNLP/cosine-discounting Public

    Jupyter Notebook 2

  6. MSc-Data-Science-and-AI-U-of-Liverpool MSc-Data-Science-and-AI-U-of-Liverpool Public

    Jupyter Notebook