Menu

Sentiment analysis for equity trading, credit markets, and customer experience

A remote control
Table Of Contents

Share this page

Mike-McNamara
Mike McNamara
70 views

In the financial services industry, the ability to accurately assess the sentiment of spoken or written communications has significant benefits for diverse use cases including equity trading, assessing credit markets, and understanding customer experience. There is no shortage of voice and text data, which is growing exponentially, but it can be a challenge to reliably extract clear signals from this data.

Sentiment analysis offers opportunities to uncover meaningful insights embedded in quarterly earnings calls, research reports, company filings, and other sources. Using natural language processing (NLP), sentiment analysis can rapidly analyze large quantities of voice and text data, delivering insights before the currency and value of the data decline.

An AI pipeline to analyze sentiment in financial services

NetApp and SFL Scientific have developed an easy-to-implement AI pipeline that captures sentiment from text or audio, including real-time conversations. The joint solution can be quickly deployed on premises or in the cloud and trained and tailored to your specific financial services use case to extract valuable insights and improve your company’s competitive position.

By combining NetApp’s hardware and data management tools with SFL Scientific’s comprehensive data science and data engineering skills, we developed an end-to-end AI pipeline that’s applicable to financial services use cases. The pipeline uses a modular architecture, deploys quickly, and can be continuously refined to improve accuracy and satisfy the needs of your business.

In operation, voice and text data is processed through a pipeline with pretrained speech recognition and sentiment modules to determine sentiment at the sentence level. Audio data is ingested and converted to text, then each sentence is assigned a sentiment value of positive, negative, or neutral. Sentiment is scored per sentence and tracked over time, with results sent to a dashboard or stored for subsequent analysis. Results are available in near-real-time for use cases that require it. For example, if sentiment analysis shows that an earnings call turns negative during Q&A, that information could determine the next day’s trading strategy for the company’s stock. 

Workflow

Additional opportunities

This white paper describes the rapid implementation of a state-of-the-art deep learning model to detect sentiment in diverse spoken and written communications for financial services use cases. The general methodology described is applicable to a broad range of NLP and other AI challenges.

Mike McNamara

Mike McNamara is a senior leader of product and solution marketing at NetApp with 25 years of data management and data storage marketing experience. Before joining NetApp over 10 years ago, Mike worked at Adaptec, EMC and HP. Mike was a key team leader driving the launch of the industry’s first cloud-connected AI/ML solution (NetApp), unified scale-out and hybrid cloud storage system and software (NetApp), iSCSI and SAS storage system and software (Adaptec), and Fibre Channel storage system (EMC CLARiiON). In addition to his past role as marketing chairperson for the Fibre Channel Industry Association, he is a member of the Ethernet Technology Summit Conference Advisory Board, a member of the Ethernet Alliance, a regular contributor to industry journals, and a frequent speaker at events. Mike also published a book through FriesenPress titled "Scale-Out Storage - The Next Frontier in Enterprise Data Management", and was listed as a top 50 B2B product marketer to watch by Kapos.

View all Posts by Mike McNamara

Next Steps