INFINIQ VietNam | Ha Noi City, Viet Nam CTV-Annotator
September 2021 - March 2022
Project Involvement:Auto Drive Annotator
Project Description:Labeled and segmented objects in images and converted them to color format.
Key Responsibilities:
Color-coded object detection and segmentation.
Labeled objects in images and converted them to color format.
Frame work: MyCrowd
PROJECT
August 2024 - Present
Customer Segmentation through Feedback Analysis using NLP - Individual
Project Description:This project involves analyzing customer feedback using Natural Language Processing (NLP) to segment customers into distinct groups. By extracting and interpreting insights from feedback data, the project aims to identify patterns and preferences, allowing businesses to tailor their marketing strategies and improve customer satisfaction.
Key Responsibilities:
Scraped customer feedback data from the Điện Máy Xanh website using Selenium to gather insights.
Annotated the feedback data with sentiment labels (positive, negative, neutral) using Label Studio to facilitate accurate analysis.
Performed data cleaning and preprocessing to ensure high-quality input for model training.
Trained and fine-tuned the PhoBERT language model to evaluate and classify feedback sentiment effectively.
Developed a full-featured e-commerce website with CRUD operations to manage product listings and customer interactions.
Integrated the sentiment analysis model into the website, enabling real-time feedback analysis and enhancing user experience.
Prediction of apartment prices in Vietnam 2024 . - Individual
Project Description:This project focuses on predicting apartment prices in Vietnam for the year 2024 through independent variables such as location, size, features, and other relevant factors. By analyzing these variables, the project aims to provide accurate price forecasts and assist investors and buyers in making informed decisions.
Key Responsibilities:
Data Collection: Gather data on apartment prices, features, and locations from various sources including real estate websites and market reports.
Data Preprocessing: Clean and prepare the data for analysis, including handling missing values and normalizing features.
Exploratory Data Analysis (EDA): Identify patterns, trends, and correlations in the apartment price data.
Visualization: Create visualizations to illustrate price trends and distribution, aiding in the understanding of data insights.
Predictive Modeling: Develop and evaluate machine learning models (e.g., regression models) to forecast apartment prices for 2024 based on historical data.
Project Description:This project aims to predict the closing price of the FPT stock on Vietnam's stock market.
Key Responsibilities:
Overall Context: Explore the FPT stock in detail by vnStock, including:
Stock history
Financial reports
Financial ratios
Trading volume
Data Collection: Scrape closing price data for the FPT stock from the VNDirect exchange using the VNstock library.
Data Preprocessing: Clean and transform raw closing price data for efficient analysis.
Exploratory Data Analysis (EDA): Identify trends, patterns, and relationships within the closing price data of the stock.
Visualization: Create interactive charts and dashboards to monitor stock performance and closing price trends in real time.
Predictive Modeling: Use ARIMA and LSTM machine learning models, along with a bagging model, to forecast future closing prices based on historical data.
Technologies Used:VNstock library, web scraping tools,ARIMA, LSTM, Bagging, Scikit-learn, TensorFlow,Python, Pandas, Matplotlib, Seaborn, Google Colab
October 2022 - April 2023
Developed a Traffic Sign Recognition Application - Team
Project Description:Developed a system for recognizing traffic signs using computer vision.
Key Responsibilities:
Collected, processed, and annotated image data from various websites for model training.
Used YOLOv5 to train the image recognition model (Google Colab)
Implemented algorithms to enhance detection accuracy, especially for small objects, including image tiling.
Integrated the AI model into a mobile application for real-time image recognition through the camera.
Technologies Used:YOLOv5, OpenCV, Python, Data Scraping, TensorFlow, Flutter