What are text analytics?

• Text analytics can help organizations discover patterns in large unstructured data sets. Unstructured data, such as videos, photos, and audio accounts for at least 80% of your company’s data, a true blind spot for most businesses. And every second, companies around the world add data to the pile exponentially. Every second 2.5 billion emails are sent. Facebook alone generates more than 22 trillion messages on top of other social media content.
• Such widespread growth of unstructured data creates an opportunity for your enterprise to use this data and create smart, optimized experiences.
• Want to know what drives performance, why metrics change, and how to impact desired outcomes?

Text analytics, powered by natural language processing (NLP), automatically and in real-time surfaces actionable insights and provides your employees with the tools they need to pull rich insights from their massive trove of data. Also referred to as text-mining, data mining, or content mining, text analytics is a form of artificial intelligence that converts unstructured data into insights that enable businesses to discover patterns in large unstructured data sets. These actionable insights translate into improved business decisions and smarter experiences.

On 4 December 2019, IBM announced that content mining is officially available as a text-analytics and data-mining capability in Watson Discovery for Cloud Pak for Data. This text-analytics feature was announced alongside other new NLP features, including enhancements to Smart Document Understanding, and the addition of another new feature, Content Intelligence.

In summer 2020, content mining also became accessible for users of the IBM Watson Discovery Premium plan.

Learn more about Watson Discovery

How is Watson Discovery leveraging text analytics?

• It is difficult to find insights when the user lacks an understanding of what they are seeking. Unlike traditional enterprise search or search engines, where users know exactly what type of answer they are looking for, content mining focuses on proactively finding hidden insights and helps users quickly surface information using a guided navigation experience. This data mining tool uses natural language processing to understand the semantics of language and machine learning algorithms to extract potentially valuable patterns held within datasets no matter where they reside: Excel documents, PDFs, online reviews, social media, raw text, and more.
• In real time, content mining provides business intelligence by enabling you to search across documents and to create a visualization of text analytics results, relationships, anomalies, sentiment analysis, and how different elements of your content are changing over time. Your teams do not need to be experienced data scientists to use Watson Discovery’s text analysis tools. For example, car manufacturers can monitor trends within customer reviews on items such as brake systems of a specific car model. Watson Discovery understands the feedback and connects it to a specific time and event, such as a big snowfall in February.

Client Stories

• Honda is one of the world’s most innovative companies, and its engineers constantly strive to design and build the smartest, most technologically advanced products on the market. To this end, it invests billions of dollars per year in its research and development organization, Honda R&D. The R&D team, located in Honda’s test facilities, recognized new data sources such as vehicle diagnostics and telematics, smartphones, biometric sensors, and large bodies of unstructured text such as customer feedback and customer surveys. These sources hold great value potential, but the engineering team did not know how to unlock the insights hidden within these huge data sets. Today, Honda’s engineers can see beyond their test facilities to address a critical business need. The engineering team is using IBM Watson Discovery’s text analysis capability, content mining, to gain a better understanding of how cars and drivers behave in the real world. They are able to pull insights for quality assurance and customer experience purposes, which include assistance in diagnosing and repairing vehicles and detecting vehicle defects.

• Content mining also plays a key role in the digital transformation of Korean Air. Korean Air has years’ worth of historical maintenance records for the hundreds of aircraft in its fleet. But until recently, this vast amount of critical data was virtually unsearchable. That meant that maintenance technicians had to diagnose and fix issues without being able to tap into or interpret implications from valuable past learnings and courses of action. Using content mining, Watson Discovery delivered actionable insights on the root causes and solutions of issues, which enabled Korean Air to shorten its maintenance defect history analysis lead times by 90%. The maintenance employee can now see patterns of defect and failure on equipment and make preventive maintenance, allowing the company to spend more time working to keep their 25 million passengers happy.
• For Verizon, with more than 100 million customer relationships, the question is how to optimize millions of customer interactions to deliver a best-in-class experience. Providing a personalized experience requires knowing the journey of customers, understanding customers’ intent, and gathering real-time insights. Verizon used Watson Discovery’s content mining feature to create a customer experience analytics platform that provided them with predictive analytics. The automation of analyzing customer-related text data is helping Verizon “predict the customer intent by text mining unstructured data and correlations.”

Learn more about Watson Discovery’s powerful new NLP features