18340771. MACHINE LEARNING AND NATURAL LANGUAGE PROCESSING (NLP)-BASED SYSTEM FOR SYSTEM-ON-CHIP (SoC) TROUBLESHOOTING simplified abstract (QUALCOMM Incorporated)

MACHINE LEARNING AND NATURAL LANGUAGE PROCESSING (NLP)-BASED SYSTEM FOR SYSTEM-ON-CHIP (SoC) TROUBLESHOOTING

Organization Name

QUALCOMM Incorporated

Inventor(s)

Murat Cakir of San Diego CA (US)

Lindsey Makana Kostas of San Diego CA (US)

Narasimhan Narayanan of San Diego CA (US)

Jonathan Charles Brauer of Austin TX (US)

Michael Robert Manahan of Rancho Mission Viejo CA (US)

Ruitao Dou of Irvine CA (US)

Yanjia Chen of San Diego CA (US)

Pradeep Mohanan Nair of San Diego CA (US)

MACHINE LEARNING AND NATURAL LANGUAGE PROCESSING (NLP)-BASED SYSTEM FOR SYSTEM-ON-CHIP (SoC) TROUBLESHOOTING - A simplified explanation of the abstract

This abstract first appeared for US patent application 18340771 titled 'MACHINE LEARNING AND NATURAL LANGUAGE PROCESSING (NLP)-BASED SYSTEM FOR SYSTEM-ON-CHIP (SoC) TROUBLESHOOTING

Simplified Explanation

The abstract describes a method for processor-implemented IC troubleshooting using natural language processing and machine learning.

• Receiving an IC troubleshooting query from a user
• Clustering the query into semantically similar categories
• Retrieving resolution data from an expert system library
• Generating a recommendation based on the resolution data
• Outputting the recommendation to the user

Potential Applications

This technology could be applied in various industries such as electronics manufacturing, IT support, and consumer electronics for efficient troubleshooting and problem-solving.

Problems Solved

1. Streamlining the troubleshooting process for IC issues 2. Providing accurate and relevant recommendations based on user queries 3. Enhancing user experience by offering quick solutions to problems

Benefits

1. Improved efficiency in resolving IC troubleshooting queries 2. Enhanced user satisfaction with timely and accurate recommendations 3. Reduction in downtime for IC-related issues due to faster resolution times

Original Abstract Submitted

A method for processor-implemented method includes receiving an integrated circuit (IC) troubleshooting query for an IC. The IC troubleshooting query is received from a user. The method also includes performing natural language processing and machine learning to cluster the IC troubleshooting query into one of a number of semantically similar troubleshooting categories. The method further includes retrieving resolution data from an expert system library, based on a mapping between categories of user solutions and a topic of the IC troubleshooting query. The method also includes generating a recommendation in response to the IC troubleshooting query, based on the resolution data. The method outputs the recommendation to the user.