top of page

CogniPath: How Generative Ai Complement RPA To Improve Process Efficiency?

Organizations generate vast amounts of unstructured data in the form of emails, feedback forms, and other text-based documents. Extracting valuable insights from this unstructured data can be a challenging and time-consuming process. However, by combining Robotic Process Automation (RPA) with Generative AI, we are able to supercharge the accuracy from previously 71% to 97% and shortened implementation lead time from 3 months to 3.5 weeks for automating data extraction from unstructured documents and carry out the order processes in ERP system. Here's how the process works:

  1. Data Ingestion: The RPA tool ingests unstructured data sources, such as customer feedback forms, and passes the data to the Generative AI model for analysis.

  2. Information Extraction: The Generative AI model uses Natural Language Processing (NLP) techniques to analyze the unstructured data and extract relevant information such as names, dates, product names, and quantity analysis.

  3. Order Processing: The RPA tool automates the order process based on the structured data extracted by the Generative AI model.

  4. Customer Communication: The Generative AI model can be used to generate human-like responses to customer inquiries or feedback, allowing the RPA tool to respond to customers in a personalized and timely manner.

  5. The generative AI will periodically generate insight for commercial team to engage customers.

CogniPath's data scientists team has fine tuned the pre-trained model with the below techniques:

  1. Preprocessing: Before extracting information from unstructured documents, ChatGPT uses preprocessing techniques to clean and prepare the text. This can include tokenization, stemming, and stopword removal.

  2. Named Entity Recognition (NER): ChatGPT uses NER algorithms to identify and extract named entities, such as people, organizations, locations, and dates, from the preprocessed text.

  3. Part-of-Speech (POS) Tagging: POS tagging to identify the parts of speech in the text, such as nouns, verbs, and adjectives. This information can be used to identify relevant keywords and phrases.

  4. Dependency Parsing: Dependency parsing algorithms to identify the relationships between words in the text. This can help identify the subject, object, and action in a sentence, which is useful for identifying relevant information.

  5. Machine Learning: trained on a large corpus of text data, which allows it to learn patterns and relationships in the text. This enables it to understand the context of the text and extract relevant information.

CogniPath, Ai from KY & Company continues help the clients to harness the power of AI to accelerate growth and improve the organization's efficiency.

Please download the full pdf here

CogniPath Supercharge efficiency with RPA and generative AI
Download PDF • 6.83MB

About KY & Company Full-service digital transformation partner that integrates Strategy, Design, Engineering and Managed-services for Corporates & Government 

Mike Kwok Managing Director KY & Company Hong Kong Office

Raymond Kam Business Analyst KY & Company Hong Kong Office 

For general inquiry, please Email visit our web site

bottom of page