ASSESSING AND REDUCING COMPLEXITY OF TEXT DOCUMENTS Next Case Study

Customer:

A Product Suitable For Any Government With Substantial English Documentation

Business Background:

One of the day to day activities in an IT firm involves dealing with various text documents such as employee offer letters, policy documents, client related documents, HR documents etc., these documents are quite often lengthy and difficult to understand. Majority of these documents are crucial for the growth and reputation of the company and clear, easy to understand. Unfortunately, due to the complexity of these documents not all documents are understood by everyone and it is quite essential to make these documents easy to understand.

Need For Analytics:

Majority of the documents involved in the IT firm play a crucial role in the growth and reputation of the company. Apparently, these documents should be clear and easy to understand. However, defining goodness or ease of understanding a document is tricky. Nevertheless, goodness or ease of read of a text document can be defined using certain standards of writing. Plain language is one such set of writing techniques. It is, as defined by US govt., a communication your audience can understand the first time they read or hear it. Application of text mining displays the potential to automate the process of measuring the goodness of document and it also has the capability to suggest improvements.

Technical Solution:

Collated various documents from various verticals and horizontals of the company such as HR documents, Finance documents, Marketing documents, Project Management documents etc. We applied text mining techniques and built a model which exhibits two important features, they are, Auto Guidance for writing and correcting any document and Auto Analyzer to analyze the documents. Auto guidance method gives auto suggestions.

For example, if a sentence is written in a passive voice, it suggests an active voice of the sentence; if a complex word is used, it suggests an alternate commonly used word. Auto Analyzer analyses the document across various dimensions such as, design of the document, organization of content, language used, context analysis etc. In addition, the end user can view the areas for improvement and make the required modifications.

Software / Tools:

Quadratyx NLPTook Kit, Java, JSP

Outcome:

The end-to-end process has been executed successfully and a model is deployed successfully delivered and was received.