Natural Language Processing

Fintechs that use Natural Language Processing aims to understand unstructured text by and voice using machine learning and artificial intelligence respond appropriately – like chatbots, extracting information from conversations, sentiment analysis and automatic summarisation

In its simplest form, Natural Language Processing (“NLP”) is the ability of a computer program to understand human speech as it is usually spoken. It is different to the older versions of speech recognition and text to speech programs that have been well known for some years but these were essentially rules based where either text or speech are compared to libraries of text or speech wave patterns.

Computers have used formal languages such as XML, PHP, SQL that have no ambiguity possible. Development of linguistic frameworks like Meaning-Text Theory, distributed computing and neural networks allow computers to process natural language and start understanding the meaning underneath the human language.

NLP enables computer programs to understand unstructured text by using machine learning and artificial intelligence to make inferences and provide context to language, just as human brains do. It is a tool for uncovering and analyzing the "signals" buried in unstructured data. Companies can then gain a deeper understanding of public perception around their products, services and brand—as well as those of their competitors.

There are many obvious use cases like voice activated search or translation of languages but far fewer have been used in financial services.

The main use cases in financial services are generally:

1. Gather real-time intelligence on specific stocks- create a real-time alert if analysts upgrade or downgrade a stock, thereby offering clients a valuable trading edge
2. Response handling – computer programs that interact with clients to listen and understand queries and suggest appropriate answers
3. Anticipate client concerns- discover and parse customer sentiment by monitoring social media and analyzing conversations about their services and policies.
4. Sentiment analysis – analyse conversations (e.g. from social media) and extract sentiment around stock prices, for example
5. Automatic summarisation – ability for a program to analyse monthly fund performance, for example – and speak the commentaries in a natural manner
6. Information extraction – analyse information from voice or text sources and provide context for quicker resolution of issues

Some solutions

Response handling

• Natural Language Processing for customers to interact with intelligent robotic assistants
• Natural language interaction by clients with automated system that answers voice queries
• NLP system to better understand context of speech and text and use machine learning to respond

Sentiment analysis

• Extract stock related conversations from social media and rates financial bloggers

Automatic summarisation

• Analyses Natural Language Processing on credit applications to predict default

Anticipate client concerns

• Allows clients to ask questions in free speech and uses NLP to deliver answers using AI
• Answers clients queries using AI and process their queries using NLP

Information extraction

• Comprehensive property database with prices, market trends and signals gained from analysing natural language processing

• Prev
• Next