What You Need To Know about Text Analytics

You have enough to worry about. You know your industry inside and out. You know your products and services and how they compare with the competition’s strengths and weakness. In business, you have to be an expert in a range of topics. What you don’t need to worry about is the in’s and out’s of every technology, algorithm and software program.

This is especially true of an inherently complex technology such as natural language processing. As a business owner you have enough to worry about. Do you really have time to understand morphological segmentation?Text analytics should just be another tool in your toolbox to achieve your business ends. The only thing you need to know is what problems you have that can be solved by Natural Language Processing. Anaphoric referencing? Don’t worry about it. We have it covered and anything else you might need from language technology.

Text Analytics

What you do need to know about text analytics?

Text analytics goes by many names: natural language processing, NLP, text analysis, text mining, computational linguistics. There are shades of difference in these terms but let the boffins work that out. What you need to know is that these terms describe a variety of algorithms and technology that is able to process raw text written in a human language (often referred to as a natural language) to provide enriched text. That enrichment could mean a number of things:

  • Categorization. The categorization of the text according to themes, categories or a taxonomy.
  • Topic Extraction. The identification of the key named entities and concepts being talked about in the text such as people, place, organizations and brands.
  • Sentiment Analysis. The analysis of whether the text is talking about those concepts in a positive or negative light.

How these can help you in your business

Categorization (aka Classification). The manual task of categorizing text is laborious, expensive and, in a world where we are creating more information in one day than human’s created in the last two millennia, arguably impossible. Auto-categorization of text helps users navigate and find the content they are looking for. Categories help ensure that the right content goes to the right person at the right time. It enables the possibility of creating governance mechanisms over large swathes of content to ensure legal or organizational compliance, access privileges, retention policies, etc. All this is achieved through inexpensive and automated means with a proven track record of success.

Topic Extraction. Where categorizations is about pinning an entire text to a single category, topic extraction is about finding the constituent named entities and concepts. Another way to put it is that it finds all the ‘things’ the text is about. This provides key insights to understand the meaning, provenance and similarity of documents. You are getting a more robust and complex view of the content. This enables information discovery, faceted searching and clustering of documents according to the concepts being described. It also enables the discovery of relationships between concepts. Using topic extraction over a period of time, it is possible to develop early warning systems for trending concepts.

Sentiment Analysis. Sentiment analysis is a key functionality for many organizations interested in text analytics. As businesses, we need to be responsive to our customer’s needs. Speed and accuracy is essential. Sentiment analysis enables you to not only understand what your customers, employees and users are talking about but whether what they are saying is positive or negative. You can see why this is important. Two statements, ‘The quality at company X is terrible.’ vs. ‘The quality at company X is great’ have very different ‘actionable insights for company X. This is just the start of the complexity.

For sentiment analysis, context is important. Whether the phrase ‘interest rates are falling’ is positive and negative depends strongly on the context. Are we talking about ‘loans’ or ‘savings account’? Is it in the context of a specific banks performance or with respect to the overall economy? The technology capable of handling that complexity is the result of decades of research. It makes no sense for a company to reinvent that wheel, especially considering that MeaningCloud provides that technology through a number of easy to use APIs and user interfaces.

With MeaningCloud, You Become an Instant Expert in Text Analytics

MeaningCloud is continuously refining its text analytical engine to provide its functionality to address current and emerging business needs. This functionality (e.g. categorization, entity extraction, sentiment analysis) is made available to users via high-level APIs. We further support developers through plugins and SDKs for a variety of environments and languages. For non-technical users we have an add-in for Excel that makes complex language analysis tasks a simple button click. MeaningCloud is confident that the plug-in-play approach to text analysis will quickly demonstrate it value. To encourage you to see for yourself, MeaningCloud offers a free 40,000 requests per month plan. Fee-based plans start at $99 a month.


Leave a Reply

Your email address will not be published. Required fields are marked *