f4ruk-seker/data-tagger

Project Portfolio: Data Marking Service

Project Name: Comment Classification and Data Cleansing Tool

Project Description:

Data Marking Service is a service designed to classify comments collected from various data sources and analyze comments specifically for service-oriented businesses such as restaurants, cafes, and bakeries. This service retrieves comments from data sources through scrapers that automatically pull comments and then converts them into prepared data for machine learning algorithms. Additionally, with data cleansing and editing tools, it plays a significant role in the preparation and export processes of the data.

Project Stages and Features:

Data Collection: Scrapers are developed to automatically retrieve comments from various platforms related to businesses such as restaurants, cafes, and bakeries (e.g., social media, review sites). Data Classification: Collected comments are classified according to the type of business (restaurant, cafe, bakery, etc.). This step performs the assignment of comments to specific categories by analyzing their content. Data Cleansing and Editing: The received comment data is prepared with operations such as cleaning unnecessary characters, proper formatting, and other data editing processes. This step ensures more accurate results from machine learning models. Machine Learning Model Preparation: The cleaned data is used as training data for machine learning algorithms. Appropriate models such as classification or sentiment analysis are developed to achieve accurate results. Evaluation of Results: Developed machine learning models are tested on real-world data and their performances are evaluated. Automation and Integration: In the final stage of the project, the developed service is automated to work automatically and integrated with other systems. This ensures the continuous and efficient use of the service.

Technologies and Tools:

Python programming language Web scraping tools (BeautifulSoup, Scrapy) Natural Language Processing (NLP) libraries (NLTK, spaCy) Machine Learning libraries (scikit-learn, TensorFlow, PyTorch) Database management systems (MySQL, MongoDB) Web application development tools (Flask, Django) This project is designed to help businesses analyze customer feedback more effectively and improve their services. Additionally, by enhancing data quality with data cleansing and editing tools, it contributes to more reliable results from machine learning models.

• login with token