Strategies of automated annotation of online media content

Abstract

We approach a novel method to enrich a text data with social media that implicitly reference certain part of it. The aim is to provide automated annotation on large sized Twitter data by generating automatic hashtags so that user will be able to nd important link and organized information quickly in one place. We provide automated annotation method for user to do automatically recommended annotation on their tweets. User are either allowed recommended annotation by accepting hashtags or not. For that we also describe how to use Apache Stanbol for providing annotation to user with use of natural language processing. For this case we are also including in the checking a set of keywords to re ne the incoming tweets. Each tweets are annotated with signi cant Freebase/DBpedia entities with three tuples (people, Organization, Place) likewise for each link contained in the tweets a same analysis is performed by fetching URL and pull out any HTML tag with Apache Stanbol facade. And also con gure Geo-location in Apache Stanbol for nd any location.