Text Mining/ Sentiment Analysis
What comes to mind when you hear text mining? Or sentiment analysis? When I first heard of these terms, I immediately thought of little people working in mines and the emoticons on our smartphone keyboards. Text mining is the process of examining huge collections of data/written resources to generate new information. Using this new information, the unstructured text is turned into structured data so it can be further analyzed. Like the little people working in the mines, we use technology to sift through these huge masses of data and identify facts, relationships, and other assertions that would’ve originally been buried. Along with text mining, we also use sentiment analysis to figure out what is being said. Sentiment analysis is also a type of process to figure out if these pieces of data are in a positive, negative, or neutral tones. Like the emoticons on our keyboards, we can figure out the tone of the particular topic. These type of analyses help uncover the true meaning of these unstructured data to turn it into structured information we can use to bridge the gap between people and technology.
Text mining and sentiment analysis both tie into the topics we covered so far in MIS 2502 because they’re both forms of data analytics. A topic that would correlate to text mining and sentiment analysis would be association rules. By using text mining, we can identify relationships between different objects and help predict different buying situations. Like association rules, we can figure out what would the customer most likely buy in conjunction with what’s currently in their cart. For example, the Target case would be a great example of text mining, through google searches, Target figured out the teen girl was pregnant before her parents did. From what she was searching and identifying the correlation of her searches and pregnant woman’s, they figured out what coupons to send her. Using sentiment analysis, we can figure out what is the customer’s reaction to a restaurant as well. By using sentiment analysis, we can figure out the tone of their yelp review, if it was a positive or negative reaction through their uses of words. This would also help improve market research. Extracting certain words from that review, we can also figure out why someone would or wouldn’t like the restaurant. Data analytics in the business aspect is used to improve buying situations and help steer the customers to where they’ll most like to go. These analyses build off what we learned so far in class because they’re advanced and modern ways to make use of the data collected.
Cited Sources Bibliography
- Prabowo, Rudy, and Mike Thelwall. “Sentiment Analysis: A Combined Approach.” Journal of Informetrics, Elsevier, 6 Mar. 2009, www.sciencedirect.com/science/article/pii/S1751157709000108.
- Kobayashi, Vladimer B. Kobayashi B., et al. “Text Mining in Organizational Research.” Organizational Research Methods, Sage Journal, journals.sagepub.com/doi/abs/10.1177/1094428117722619.