Sinavizcisi
About the Project
Sinavizcisi.com
A Platform That Connects YKS Data with AI
About the Platform
Sinavizcisi.com is a comprehensive web platform that collects and analyzes YKS placement data
and helps students make data-driven decisions during the preference process.
But this project is much more than a website for me; the website is only a tool to present the data.
The real journey includes extensive data collection before the site existed, cleaning and analyzing
the data, applying open-source LLMs to parts of it, and creating brand-new insights that cannot be
found anywhere else.
This project became my biggest work that combines my experience in web development, data science,
and AI. It was extremely valuable for me both in experience and in return. On the visibility side,
we gained more momentum than expected in a short time.
Feedback from the site included requests to use our data in other projects,
direct requests for our datasets, and especially a dean contacting us directly.
Our data was also used in a district governorate presentation, showing the concrete impact of this work.
Performance and Visibility
We first launched the site on July 18, 2025. About two weeks later, when Google began indexing our pages, we saw a sharp rise without any ads, reaching over 8,000 page views per day. This growth came purely from organic search traffic.
While views dropped over time, traffic stayed in the 2-3 thousand per day range during exam season. After the exam period, we had a stable six-month window between October 2025 and March 2026 and achieved over 200,000 page views. Sustaining visibility for about 10 months without advertising shows that this project meets a real need and that we present data in a useful, clear way. Google Search Console data also confirms strong search performance: in August 2025 alone, we had 700,000 impressions. Achieving an average position of 14.9 in such a large search volume indicates above-average ranking. We ranked high on many high school pages; at our peak, for the search "Galatasaray High School", we appeared right after the school's own page. While this visibility has softened for some school names, we still rank around 4th-5th for searches like "Galatasaray High School placement results".
Achievements & Impact
Technical Details
System Architecture
Data Layer
- YOK Atlas data (high schools, departments, quotas, base scores)
- Yearly statistics for high schools and departments
- User reviews and social data collected from the web
- Database: PostgreSQL (for relational data)
- Data cleaning and checks: Pandas
Backend
- Django + Django REST Framework
- Async data fetching: aiohttp + asyncio
- API services: high schools, departments, quotas, base scores, reviews
- Security: rate limiting, caching, query optimization
AI / Analytics
- Review analysis: sentiment and content analysis of user reviews
- Results are served to the frontend via API
Frontend
- Technologies: HTML, CSS, JavaScript
- University and high school filtering and listings
- Charts and tables (Chart.js or similar)
- Responsive design
- Performance: minimal and fast, data loading with lazy loading
Data Flow & Processing
Data Collection
Automated data collection from multiple sources with Selenium, Playwright, and Requests
Normalization
Cleaning and standardization with Pandas
AI Processing
Analysis and prediction with Transformers
Serving
Data delivery to users via the Django API
Other Projects
Take a look at the other projects I built
YokAPI
A data layer that normalizes YOK Atlas data and serves it through a single API. It standardizes YOK Atlas's scattered, …
EBA Score Bot
A desktop bot with a GUI that automates earning points on EBA.
Turkish University Department Data (2019-2024)
A comprehensive dataset that cleans and standardizes 2019-2024 undergraduate and associate program data collected from YOK Atlas and OSYM.