Sentiment Analysis for Public Opinion Mining

May 30, 2018 | Sentiment Analysis, Social Media Analysis

Sentiment Analysis for Public Opinion Mining

Social media has made available an abundance of unstructured data and semi-structured data of great value to organizations that monitor, analyze, and forecast public opinion. Organizations such as media companies, think tanks, and polling organizations are interested in measuring public opinion in a variety of areas such as confidence in public institutions, support for proposed policies/legislation, shifting public opinion towards hot button issues, and public approval of political leaders.

The challenge is how to transform the staggering and exponentially growing volumes of unstructured data into usable information and actionable insights. Facebook alone produces vast amounts of data every second with 2.19 billion monthly active users worldwide. Manual analysis would be too slow and is simply impractical.

Sentiment Analysis Software for Opinion Mining

Sentiment Analysis software can handle this volume, variety, and velocity of data to detect and measure opinions, intent, likes, and dislikes in text.

Compared to labor-intensive and time-consuming phone surveys and public polls, Sentiment Analysis can examine a much broader sample of the population for a fraction of the cost, operate in near-real time, and capture new topics as they arise, unlike surveys, which are limited to known topics.

Sentiment Analysis can also feed predictive models to anticipate public reaction and forecast events such as election results or social unrest. Public sentiment is indeed a precursor to many such events.

Why Use NetOwl’s Advanced Sentiment Analysis?

Compared to other software, NetOwl’s AI-based Sentiment Analysis goes beyond positive and negative sentiment. Through its Entity- and Aspect-based sentiment analysis, NetOwl is able to detect the objects of sentiments (e.g., a politician, a political party, a policy) and the specific aspects of those objects (e.g., a politician’s character, a political party’s stance on an issue) that the sentiments are about. For instance, your average sentiment analysis software may detect negative sentiment around a political leader or policy but may not be able to determine what specific aspects the negative sentiment is related to.

Furthermore, NetOwl’s Sentiment Analysis offers a more refined sentiment ontology to distinguish, for example, between a complaint and a threat to boycott. NetOwl is also fast, scalable, and consistent making it possible to process massive amounts of data in near-real time without getting “tired” and making different judgements for the same text.

Finally, NetOwl normalizes the identified sentiments and entities in an intelligent way so that they can be aggregated and quantified to provide the overall view over a large collection of content.

NetOwl’s Sentiment Analysis dashboard provides multiple interactive charts to slice and dice sentiment data as desired. For any entity of interest (e.g., a proposed policy), the user can see the breakdown of public opinion, the top positive and negative sentiments, the specific aspects that are the object of those sentiments, and, if desired, drill down to the source text for inspection and further analysis. Other useful charts show sentiment evolution over time.  Using location metadata and automated geocoding of mentioned place entities, NetOwl’s Sentiment Analysis dashboard plots sentiment on a map to show, for instance, hot spots of support or disapproval.

NetOwl’s Sentiment Analysis software can greatly help organizations interested in measuring and better understanding public opinion.