Project description

1. Clustering algorithms:

• Isolate articles in English and Russian

• Isolate news articles

• Group articles into categories

• Group articles into threads

2. Analyze, store and index articles.

3. Return news threads for a particular period, sorted by relative importance.

Full description (ENG): https://contest.com/docs/data_clustering2/

How to start

Stage 0. Install

1. Install deb packages from deb-packages.txt

2. Install python modules:

pymodules.sh

Stage 1. Clustering

1.1. Grouping by language

tgnews languages <source_dir>

1.2. Isolating news articles

tgnews news <source_dir>

1.3. Grouping by category

tgnews categories <source_dir>

1.4. Grouping similar news into threads

tgnews threads <source_dir>

Stage 2. Indexing and Ranking

2.0. Start server

tgnews server <port>

2.1. Indexing

HTTP request:

PUT /article.html HTTP/1.1
Content-Type: text/html
Cache-Control: max-age=<seconds>
Content-Length: 9

<content>

2.2. Removing from the index

HTTP request:

DELETE /article.html HTTP/1.1

2.3. Thread ranking

HTTP request:

GET /threads?period=<period>&lang_code=<lang_code>&category=<category> HTTP/1.1