Sentiment analysis of Hillary Clinton and Donald Trump. | Image: OpenText
The power of big data is undeniable. Yet for most people understanding the raw numbers produced by big data can be like trying to understand a foreign language. Big data's big promise to provide deep insight about behavior and relationships is worthless without proper translation tools.
Collecting data is vital for modern SMBs and enterprise companies, but deciphering how raw numbers can lead to better decision-making is just as important. OpenText is a platform that harvests, stores, and makes massive piles of information understandable and relevant to the layperson.
READ: Get ready for big data's wild ride (Tech Pro Research story)
OpenText crunches numbers to help users understand sentiment of social media content by performing "unstructured data analysis methodology," said Mark Gamble, OpenText's Director of Technical Marketing. The platform harvests data from a variety of sources ranging from social media sites to traditional news websites. Using an internal tool called InfoFusion, OpenText processes the raw data—connections between people and names, places, and popular topics—to derive sentiment. Users can define variables, and mountains of information is then simplified and displayed on human-readable dashboards using a JavaScript API.
The cloud-based analytics tool is used by enterprise and SMBs to test theories about products and markets. Visualizing big buckets of information using relatable language and simple charts helps managers make better data-driven decisions, said Gamble.
Campaign media coverage. | Image: OpenText
During the election cycle TechRepublic is reporting on how business can learn from how presidential campaigns use big data, traditional, and social media. The OpenText Election Tracker tool allowed us to monitor and learn about the relationship between media coverage and individual campaigns.
Here are the seven biggest campaign media trends of the primary season.
- In the 72 hours preceding this week's primary news outlets wrote 4,574 stories that discuss the election or candidates. Over 55% of those stories were had a negative tone, according to the sentiment analysis tool.
- 'Ohio' was the most popular state keyword mentioned by media companies, and 'delegate,' 'Kasich,' and 'voters' were the top trending topics.
- Media mentions of Ted Cruz dropped significantly just prior to the March 15th election. In fact, Cruz was the least mentioned of all major candidates. The last time he received this little press was August 30, 2015. Unsurprisingly, Donald Trump was mentioned by the media twice as often as any other candidate.
- On primary day Hillary Clinton received only 390 media mentions, and nearly 20% of those mentions contained negative sentiment. Approximately 30% of Donald Trump's media mentions were negative.
- Sentiment trends for Clinton and Trump shifted from negative to neutral just after the March 15th primary.
- Over the course of the campaign the press and candidates have flirted with a number of issues, but 'crime' and 'foreign policy' consistently remained the top two topics.
- The media companies covering the campaign most heavily in the 72 hours before the primary were Business Insider, Mashable, and the Huffington Post.
Topical media mentions. | Image: OpenText
Over the course of the campaign we will continue to perform simple data analysis. We hope to uncover unique insights and find ways business can benefit from campaign innovations.
If you're a data scientist, social media professional, or inquisitive TechRepublic reader we'd love your ideas on how to inspect campaign social media data. Please leave a comment below or ping us on Twitter @TechRepublic.
Read more
Source: http://techrepublic.com.feedsportal.com/c/35463/f/670841/s/4e5cc31a/sc/15/l/0L0Stechrepublic0N0Carticle0Celection0Etech0Ebig0Edata0Eidentifies0Ethe0E70Ebiggest0Emedia0Etrends0Eof0Ethe0Epresidential0Eelection0C0Tftag0FRSS56d97e7/story01.htm