RULE-BASED NLP REVEALS THE STRUCTURE OF A SENTENCE
After an open-ended customer comment is run through of Rule-Based Natural Language Processing analysis , we have structured data that can be used to extract the meaning (Topic) and emotion (Sentiment) from a sentence.
- what is the base form for each word (morphology),
- what part of speech each keyword or phrase is (noun, verb etc.)
- keywords and phrases in a sentence, and
- the syntactic and semantic rules of each language.
POOLING KEYWORDS TO CATEGORIES IS HARD WORK
But knowing the structure isn't enough. You need to come up with a way to measure open-ended customer feedback. What you need to do is pool relevant keywords into Topics. This requires a large dataset and many years of work. Best in class CX text analysis solutions use a combination of Rule-based NLP and artificial intelligence to map keywords into Topics automatically and accurately.
CX TEXT ANALYSIS ISN'T JUST ABOUT WHAT BUT ALSO ABOUT CONTEXT
It is difficult to extract the context in which a keyword refers to. For example, when it comes to air travel, it is important to know whether a sandwich is bought at the airport or in the airplane. Rule-based NLP combined with contextual keyword pooling does this well. What I mean with contextual pooling is that a keyword or phrase is mapped to different Topics based on touchpoint context.
SENTIMENT ANALYSIS REQUIRES A LARGE SET OF RULES
Most sentiment analysis tools are based on a list of bad and good words. This doesn't work. Language is ambiguous: there are just too many ways to express emotions. Sentiment analysis must use a large set of grammatical rules about what words or combinations of words mean even when the words are located far from each other. More sophisticated solutions even have rules about irony and sarcasm.
RULE-BASED CUSTOMER EXPERIENCE ANALYSIS TURNS UNSTRUCTURED TEXT INTO STATISTICAL INFORMATION
Knowing the Topic, the context in which a Topic comes up and the Topic levelc sentiment turns open-text feedback into statistical information. This information can be used to detect patterns. These patterns tell what drives or eats your customer loyalty. Acting on this information improves your company's bottom line!