News Search Engine and Story Representation (Nodejs, javascript, jQuery, Java, Apache Solr)

• Reuters RCV 1 and NYT (2007) used as initial index database.
• Focused web crawl to retrieve and index recent news using news API for eg. Guardian.
• Timeline.js to represent data in chronological summarization.
• Implemented Adaptive LDA(Latent Dirichlet Allocation) to cluster and classify news data as topic summarization. Number of topic clusters is made adaptive to news data.
• LexRank algorithm to build a coherent news story through extraction.

Website url: http://bit.ly/newsproj

Name		Name	Last commit message	Last commit date
Latest commit History 97 Commits
aay-hjc-ui		aay-hjc-ui
bin		bin
css		css
html		html
java/org/kb		java/org/kb
js		js
module_xmltohtml		module_xmltohtml
node_modules		node_modules
public		public
routes		routes
scoring_model		scoring_model
solr		solr
views		views
README.md		README.md
app.js		app.js
indexer.jar		indexer.jar
message.txt		message.txt
package.json		package.json
test.jar		test.jar
test1.jar		test1.jar
timeLine.json		timeLine.json

mimanshujain/News-Summarization

Folders and files

Latest commit

History

Repository files navigation

News Search Engine and Story Representation (Nodejs, javascript, jQuery, Java, Apache Solr)

Website url: http://bit.ly/newsproj

About

Resources

Stars

Watchers

Forks

Languages