How Artificial Intelligence makes RPA smarter: two use cases

RPA-automation-computer-robot-tools and statistics

Artificial Intelligence and RPA

Many organizations could be gaining huge operational efficiencies if they combined Artificial Intelligence and RPA (Robotic Process Automation).

In a previous post (The leading role of Natural Language Processing in Robotic Process Automation) we introduced the subject of NLP in RPA. In this post, we are seeing two use cases where Natural Language Processing (also known as Text Analytics) integrated with RPA/BPM software suites, is mature enough to solve typical insight extraction problems, conveniently and cost-effectively.

Definition of RPA

Robotic Process Automation refers to the automation of repetitive tasks. RPA executes routine tasks in a fraction of the time it takes any human to do it, and without the risk for human error. 

RPA uses software robots to carry out actions such as scraping, data aggregation, data cleaning, and communicative tasks with other applications and people to execute repetitive work.

It allows businesses to maintain low costs while giving their workers increased opportunity to tackle other needs within their organizations.

RPA, can, by example, take the repeated tasks needed to keep a CRM and automate them to guarantee that the data in the corporate CRM is synchronized with business ERP-like applications (SAP, SAGE, Oracle NetSuite, etc).

RPA can also automate a number of tedious and time-consuming email-related tasks. Consider help desk requests. They can be easily integrated into your automated email processes.

Primary companies in the RPA market include Automation AnywhereBlue PrismUIPathBe InformedJacada and Jidoka.

AI and Natural Language Processing

Artificial Intelligence  makes it possible for machines to mimic humans in functions such as learning and problem-solving. AI today is comprised of multiple disciplines, such as machine learning (ML), Natural Language Processing (NLP) and many others. Artificial intelligence algorithms are designed to make decisions, often using real-time data.

Natural Language Processing (NLP) is the “ability that allows machines to understand and interpret human language the way it is written or spoken”.

The objective of NLP is to make computers/machines as intelligent as human beings in language comprehension.

Smarter RPA with NLP

RPA technologies cannot adjust their procedures without human intervention. However, RPA may become a suitable step for incorporating more sophisticated cognitive technologies such as Natural Language Processing.

Combined with RPA, AI can dramatically improve the way companies address business processes and equally, can drive operational efficiencies.  “Intelligent Automation” is the name some give to this RPA enhancement.

Most enterprise processes are manual and repetitive, and, according to some experts, more than 70% can be automated using RPA.  Processes that require human judgment can also be automated by 15–20%.

Let us take a look at a couple of examples:

Use case 1: Automatic Classification of Documents

Documents being classified

Organizing documents consistently is a complex task that requires specialized human resources. Automatic classification opens a new range of possibilities which include both total automatization and support tools that reduce time and improve the quality of manual tagging processes and obtain more consistent results, at a faster pace and at a lower cost.

An NLP-powered RPA solution such as Automatic Document Management can become more efficient over time. As more documents are processed, the solution learns how to effectively manage variations. See the MeaningCloud blog-post about Automatic Document Classification for more details. 

Beyond classification, Automatic Document Management can recognize the arrival of a new document. For example,  in an invoice, the name of the supplier would be identified, which would then trigger an action in Accounts Payable, without the intervention of the bookkeeper. The process involves AI technologies such as optical character recognition (OCR), Neural Networks (a subset of machine learning) and natural language processing (NLP).

Use case 2: Contracts, from email to ERP

Legal documentation (contracts, sentences,  agreements,…) is a typical example of unstructured content.  Law firms and corporate legal departments can benefit from tools that allowed them to extract complex data such as parties in a contract, the terms of a particular clause, those affected by a legal procedure, and how they are affected. They could analyze, find connections, and understand legal documents better.

By example, RPA software can automatically extract the content of a contract that arrives as an attachment to the inbox of a law firm. Then, the RPA will take it to an NLP tool such as MeaningCloud’s deep categorization API, where it will extract complex data such as parties in the contract, the terms of a particular clause, those affected by a legal procedure, and how they are affected. It can detect and identify their clauses and other relevant parts such as Title, Parties, Date, Term, Assignment, Change of Control, Audit, Governing Law, Force Majeure, Indemnification, Limitation of Liability, etc.

Then the RPA software will take this information and insert it automatically to the ERP.

Wrapping it up, Robotic Process Automation, combined with Natural Language Processing, can dramatically improve the way companies do business and drive significant operational efficiencies.


About Eduardo Valencia

Data-driven by instinct and by learning. I began my career as a linguist. Some things stay with you to the grave. Since 1995, I have been leading teams and companies that have successfully developed and distributed hundreds of successful data, analytics and ICT projects.

Leave a Reply

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

*
*