Every word spoken by a customer holds the power to inform, alert, and guide companies in their quest to better understand them. Voice analytics is instrumental for companies because it transforms unstructured customer-centric voice data into valuable business insights. Your customers are already saying everything you need to know, and speech analytics enables your business to understand and act on these key learnings.
Among the many expanding technologies used within speech analytic solutions, one innovation is Named Entity Recognition (NER), also known as entity extraction, chunking, or identification.
Named Entity Recognition (NER) is a form of natural language processing (NLP) where named entities are detected, and then classified into specific categories to better make sense of the data. These entities can include things like names, products, dates, and even companies — all key pieces of information to understand what is being talked about in your customer conversations. NER trains AI models how to discern simple words in relation to various categories of topics that we as humans understand easily. For example in the following sentence: “Angela & Chase have been living in Paris for the past two years to learn French,” the entities that we pick up on are:
- People: Angela + Chase
- Place: Paris
- Date: Two years
- Language: French
As human beings, we can quickly understand what each of these words means in the sentence because we understand context. With NLP, NER trains AI models to automate their understanding in the same way — in other words, it helps them think more like us.
VoiceBase is implementing 18 specific entities for classifying your customer voice data:
- PERSON: People, including fictional ones
- NORP: Nationalities or religious or political groups
- FAC: Building, airports, highways, bridges, etc
- ORG: Companies, agencies, institutions, etc
- GPE: Countries, cities, states
- LOC: Non-GPE locations, mountain ranges, bodies of water
- PRODUCT: Objects, vehicles, foods, etc (Not services)
- EVENT: named hurricanes, battles, wars, sports events, etc
- WORK_OF_ART: Titles of books, songs, etc
- LAW: Names documents made into laws
- LANGUAGE: Any named language
- DATE: Absolute or relative dates or periods
- TIME: Times smaller than a day
- PERCENT: Percentage, including “%”
- MONEY: Monetary values, including unit
- QUANTITY: Measurements, as of weight or distance
- ORDINAL: “first,” “second,” etc.
- CARDINAL: Numerals that do not fall under another type
To apply entity extraction to your voice data today, talk to your AE or reach out to support@voicebase.