blog banner

Self learning search system for Cisco

Introduction

The client is a worldwide networking and telecommunications equipment leader, founded and headquartered in San Jose, California. The enterprise develops, manufactures, and sells networking hardware, telecommunications equipment, and other high-technology services and products. The company’s revenue is over 50 billion yearly. The leadership understood the complexity and volatility of their product demand and customers purchasing patterns. Looking to improve their enterprise search, the client reached out to Pluto7 and Google Cloud for a solution that would significantly impact their revenue and profits.

Why Pluto 7 

The client looked to Pluto7 because of their proven industry and domain experience at the enterprise level for hi-technology companies. Pluto7 brought not only domain expertise but also a deep understanding of AI, ML, and Google Cloud. The client wanted to partner with a company that would drive innovation for business transformation. As a Google Cloud premier partner, Pluto7 offers demand planning and supply chain planning solutions that are industry proven and highly accurate.

Objectives

  • To build an enterprise search platform that can help in searching the data effortlessly and quickly find the information from anywhere 
  • To develop a Product Search solution for its stakeholders to provide up-to-date and accurate answers from Cisco’s internal data repositories as opposed to external data sources
  • Cisco wants to enable self-learning in the future state solution so that it can answer complex queries and enable new insights over a period of time

Current State

  • Cisco is using Cisco Answering Machine (CAM), which does abstract text summarization.

Solution 

  • Digitization of the existing documents and answer complex questions based on the data retrieved from the doc 
  • Integration of data from different sources (structured and unstructured) to answer complex questions 
  • Creation of relationships between the data retrieved from docs and structured sources

For more information
www.pluto7.com/success-stories

industry image

Industry High-tech

Organization Name: Cisco

Challenges

  • Limited intelligent self learning system and therefore cannot scale well or continue to learn.
  • Does not automatically capture content changes and works well only with structured data.
  • Infrastructure scaling challenges as data needs increase

Results

  • Develop a search solution that covers the knowledge base with relevant available information from their repositories so that their customers, partners, and internal users can have a rich search experience helping answer more complex questions compared to the capability of the current system.
  • Developed a solution that can provide the desired search experience to the end users by leveraging AI/ML.
  • Develop an NLP engine with a joint understanding of unstructured and structured data.
  • Developed a mechanism that can forge the relationship between the different entities (products, properties, etc.) based on the expected answers provided by the solution.

Products Used

  • Google Cloud Platform
  • Google BigQuery
  • Dialogue flow
  • Cloud Vision
  • Google App Engine
  • Cloud Natural Language
  • Cloud AutoML NL
  • Cloud Translation
  • Google Cloud Search API