Online News Article Temporal Phrase Extraction for Causal Linking

Abstract

This research proposes a method to detect temporal links between news articles using phrase extraction and position analysis. The words from the information feed were split into tokens using special character as the delimiters. The words, symbols and tokens were marked by the type of word such as noun, verb, adjective, preposition and adverb. The position of each word and its length and distance from the verb on the either side of previous 2 and next 2 sentences was extracted as a feature. The time based relations between several sentences using noun as the root and verbs as the action event was stored. The link length and hop count was measured and the connected tree was stored as a decision model. The prediction model was evaluated using time based questions and event series queries. The results showed 67% accuracy in terms of correct event sequencing and recall rate. In the next step emotion based classification of the news article event actions was performed to test whether the weighted graph linking based on phrase position properties (length, count, location) method could be used for sentiment/emotion analysis with fear and happiness as the class labels. This evaluation showed clear thresholds in case of specific event-event and event-action pairs and their temporal relation type.

Authors and Affiliations

Amol S Patwardhan

Keywords

Related Articles

Review On Digital Filter Design Techniques

Measurement Noise Elimination is considered as one of the most important problem in signal processing. Several solutions have been proposed by many researchers using different techniques. Some of them have been successf...

An Introduction to Optical Character Recognition

This article is all about the development of OCR usingMATLAB. It describes all from segmentation toclassification. To understand how an experimental approach was chosen, for example, the result got affected by the use o...

slugA Novel Approach for the Development of Metadata Editors in Web-based Applications

Metadata is the data about data which describes how and when particular data was collected and how the data should be formatted. Metadata Standards that support a number of defined functions, and specify elements which...

slugMULTI-METHOD APPROACH AND FORMAL MATHEMATICAL MODEL FOR OWNER SHIP SERVICE MANAGEMENT ON CLOUD COMPUTING COST

The use of Cloud Computing Services appears to offer significant cost advantages. Particularly start-up companies benefit from these advantages, since frequently they do not operate an internal IT infrastructure. We fou...

How to Mitigate the Impact of Supply Chain Disruptions and Risks?

In this dynamic environment, supply chain disruptions are increasing and leading to more pressures toward adapting and quickly responding in an agile way to that change. The purpose of this article is to explore how org...

Download PDF file
  • EP ID EP22631
  • DOI -
  • Views 207
  • Downloads 4

How To Cite

Amol S Patwardhan (2016). Online News Article Temporal Phrase Extraction for Causal Linking. International Journal for Research in Applied Science and Engineering Technology (IJRASET), 4(9), -. https://europub.co.uk/articles/-A-22631