Sentiment Analysis with NLP Leads to More Accurate Understanding

The natural language processing (NLP) engine in the NetBase social intelligence platform reads and understands millions of social media postings every hour. For every sentence, it identifies and links the subjects, objects, verbs, adjectives, and other linguistic patterns. By analyzing this “connective tissue” within each sentence, our NLP engine can account for the complexities in language that have a huge impact on meaning. We then preserve that meaning in a special index within ConsumerBase.

For example, the sentence, “The iPhone has never been good,” is actually a negative statement in spite of the fact that it uses the word “good.” An almost identical sentence, “The iPhone has never been this good,” is positive. In the sentence, “I like using my iPhone, but I hate the way that applications work on the Droid,” the words “iPhone” and “hate” occur close together but are not associated with each other. If systems that use pattern matching were judging the sentiments expressed by the writers by the keywords “good” and “hate” alone, they would be wrong more than half the time.

Analyzing the “Connective Tissue” of a Sentence

By using the science of language to understand what is being said, NetBase NLP offers several advantages:

Very high accuracy. Our NLP engine delivers over 80% accuracy, while solutions that rely on statistical keyword-matching algorithms are less than 50% accurate because they never look beyond the context of a single word.

A system that gets the social web. It speaks standard English plus four other versions:

• Urban words or “slanguage,” for example “My new phone is sick!”
• Alternative spellings, for example “luv,” “kewl,” or “gr8”
• Abbreviations, for example “IMHO,” “ttyl”
• Common misspellings, for example “teh/the”

The NetBase NLP engine has been optimized for the specific lexicon of social media. We are constantly incorporating new rules into our own social media lexicon based on the work of our team of computational linguistics experts, ongoing testing that we do using “crowdsourced” human evaluators, and on feedback from customers.

Instant results. It has already normalized, indexed and stored up to 12 months of social media commentary in ConsumerBase. Every time you run a query, you access this vast repository of pre-indexed content so you don’t have to wait for results.