PORTFOLIO
Archived academic-era record
This archived academic-era portfolio collects undergraduate projects from a period of learning data science from near scratch. Most work here was completed in a short development window and shows how quickly new skills were picked up during BSc Computer Science studies.
On 8 December 2019, Python and simple data analytics study began, alongside an “Artificial Intelligence” elective. The projects below are products of that academic period — roughly about a year of early data-science learning as a student — not a current professional body of work.
One project in this list was done within less than 38 hours — the shortest development time among these academic-era works — starting from almost no related knowledge and learning quickly from the beginning.
Matchmaking-based Restaurant Recommendation System
on Hybrid Mobile Application
Project Due Date:
October 2020
This short video was created by: Project's UX Engineer
Small-Scale Projects
Learning by doing. These small to medium-scale project have done for learning by start self-coding and support with source from across online-course and classroom
Explain or discribe about restaurant that user want to dine out.
Keyword: NLP, Latent Factor Collaborative Filtering
Date: 7 July 2020
Quick View Document (.ipynb)
Predict probability of reviewer's statisfaction of each review.
Keywords: Natural Language Processing, Sentiment Analysis
Date: 29 June 2020
Quick View Document (Kaggle)
Recommend diner in Las Vegas with concept of blind date.
Keywords: Matrix Factorization, Collaborative Filtering
Date: 20 February 2020
Quick View Document
Sometime user perfer dining experience more than taste
So developer try to suggest the user by searching into review database
Inspriration: Yelp - Restaurant Search and Recommendation Service
Dataset: Yelp Dataset (Open data)
Instruction
User just input text of restaurant description into variable then restaurant will be recommend
Entrepreneur want to improve their business
Feedback is a key, developer want to assist by predict them all
Dataset: Amazon Musical Instruments Reviews (Kaggle Dataset)
Instruction
User just input review text into variable then satisfaction will appear
Like online dating, User perfer to eat but cannot make a decision
So developer try to suggest the user by doing blind date concept
Inspriration: Tinder - Online Dating Application
Dataset: Yelp Dataset (Open data)
Instruction
1. User start to program.
2. User begin choose their meal as user perfer.
3. Program will ask some more until it make sure what user want (about 10 meal).
4. Program will recommend restaurant that matched with user requirement.
Certificates and Awards
From proof of events participation to awards since first enroll in college (some was missing)
Archived academic-era portfolio (BSc Computer Science / ~2019–2020) — not a current professional portfolio.
This site used Font Awesome and W3.CSS as CSS Framework
© 2021 Copyright: Thanapat Pitchayarat | Firebrand Intellect
