TRAP@NCI

Martello.io: Technical Report

O'Callaghan, Adam (2017) Martello.io: Technical Report. Undergraduate thesis, Dublin, National College of Ireland.

[img]
Preview
PDF (Bachelor of Science)
Download (1MB) | Preview

Abstract

As the massive amounts of both structured and unstructured data produced daily continues to grow, it is evident that keeping abreast of current events, ideas, and trends is becoming increasingly difficult. Many press officers, journalists and media professionals resort to the old method of combing through websites, print media, and other content streams in order to get an idea of what is currently happening. However, this scattershot approach is far from optimal as key stories and information can often slip past the person’s attention. Complicating matters is non-stop, 24-hour nature of the global news cycle, whereby stories appear and disappear in quick succession. All of these factors make it close to impossible for any one person to stay aware of each breaking news story.

However, as artificial intelligence technologies continue to mature, new techniques are appearing which allow computers to track, extract, and analyse the vast array of information that is produced each day. One of these technologies makes use of a set of algorithms known as Natural Language Processing, which allows computers to analyse the content of natural human language and derive new information from it.

The application of these Natural Language Processing algorithms to online articles, news, and other media content is the central concern of the Martello.io software system. Martello will actively scrape content from the websites and newsfeeds of the top media outlets and process this information using a number of algorithms. The results of this NLP processing will allow the output of summaries and categorisation, the identification of named people, places and organisations, and the detection of the sentiment of the articles themselves in order to see if it is a positive or negative opinion.

Customers using Martello will be able to make use of a full-featured application that allows them to quickly extract information pertinent to their organisations and narrow the vast quantity of news articles down by location, category, date, or by specific people, companies and industries. Additionally, customers will be able to view the positive or negative sentiment of relevant news articles, or collections of articles, in order to quickly ascertain the prevailing sentiment about specific subjects and topics.

Item Type: Thesis (Undergraduate)
Subjects: Q Science > QA Mathematics > Electronic computers. Computer science
T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science

Q Science > QA Mathematics > Computer software
T Technology > T Technology (General) > Information Technology > Computer software
Divisions: School of Computing > Bachelor of Science (Honours) in Computing
Depositing User: CAOIMHE NI MHAICIN
Date Deposited: 25 Oct 2017 13:53
Last Modified: 25 Oct 2017 13:53
URI: http://trap.ncirl.ie/id/eprint/2642

Actions (login required)

View Item View Item