Loading [a11y]/accessibility-menu.js
From news to company networks: Co-occurrence, sentiment, and information centrality | IEEE Conference Publication | IEEE Xplore

From news to company networks: Co-occurrence, sentiment, and information centrality


Abstract:

To understand connectivity among companies in financial news, and their overall influence, we define an algorithm for ranking companies in networks of positive, negative ...Show More

Abstract:

To understand connectivity among companies in financial news, and their overall influence, we define an algorithm for ranking companies in networks of positive, negative and mixed co-occurrences. We collect a homogeneous set of financial articles from crowd sourced news to get a sentiment polarization between positive and negative news. We use this polarization to develop three types of sentiment networks by matching co-occurrences of companies in the Standard & Poor's 500 index. The entities are then ranked according to the information centrality measure and normalized by market capitalization. This text-to-network process allows qualitative and quantitative analysis of the top 25 nodes in each network on a quarterly basis over a period spanning 2011Q1 to 2016Q2.
Date of Conference: 06-09 December 2016
Date Added to IEEE Xplore: 13 February 2017
ISBN Information:
Conference Location: Athens, Greece

References

References is not available for this document.