# US 2019/0378179 A1 · Method and System for Generating Contradiction Scores for Petroleum Geoscience Entities within Text using Associative Topic Sentiment Analysis.

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2019/0378179 A1
- Document ID: US20190378179A1
- Status: Published application
- Kind: A1
- Application number: 16/005,712
- Filed: 2018-06-12
- Published: 2019-12-12
- Earliest claimed: 2018-06-12
- Assignees: ExxonMobil Upstream Research Company
- Applicants: ExxonMobil Upstream Research Company
- Inventors: Paul Hugh Cleverley
- Web page: https://patentgrep.com/patent/US20190378179A1

## CPC

- G06F 40/30: Semantic analysis
- G06N 20/00: Machine learning
- G06Q 10/40: Business processes related to social networking or social networking services
- G06Q 30/0282: Rating or review of business operators or products
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR

## Abstract

The invention is a data processing method and system for notifying petroleum geoscientists of contradictions in sentiment between petroleum geoscience topic associations in text. The data processing system makes the necessary calculations to detect petroleum geoscience associations of interest, predict their sentiment and report contradictions in sentiment. Contradictions can be calibrated with past business success and failures through time, such as exploration or operations, to create a classifier to predict future areas of opportunity and risk. In particular, the data processing system operates on any digital unstructured text derived from academic literature, company reports, web pages and other sources. This information is analyzed and utilized by the system assisting petroleum geoscientists to identify business opportunities and risks.

## Claims

### Claim 1

A computer implemented process for fine grained petroleum geoscience sentiment analysis of topic associations, the system comprising:
receiving into memory:
sentences of text, petroleum geoscience topic lexicons and a petroleum geoscience sentiment labelled training set
processing the data in the computer memory with a processor to detect petroleum geoscience topic associations
processing the data in the computer memory with a processor to assign a sentiment value to a petroleum geoscience topic association(s); and
outputting the petroleum geoscience topic association(s) and assigned sentiment value(s), by:
applying a filter around the detected petroleum geoscience topic association(s), the filter having a width defined by a number of tokens, the topics having a sequence position number in the sentence, the filter output providing an input to a statistical sentiment classifier created from the petroleum geoscience sentiment labelled training set.

### Claim 2 (depends on 1)

The computer-implemented process of claim 1 further comprising:
computation of a contradiction score based the sentiment of the petroleum geoscience topic association(s); and
outputting the petroleum geoscience topic association(s) and assigned contradiction scores, by:
calculation of the variance for the aggregated petroleum geoscience topic association sentiment classification for each of the unique petroleum geoscience topic association(s); and
incorporating an entropy based weighting function for the number of unique authors into the overall contradiction score.

### Claim 3 (depends on 2)

The computer-implemented process of claim 2 further comprising:
receiving into memory:
a labelled training set comprising the contradiction scores derived from the sentiment of petroleum geoscience topic associations where they can be linked to corresponding historical business success and failure, along with unlabeled petroleum geoscience topic associations
processing the data in computer memory using a computer processor to predict the likelihood of business success/failure for un-labelled combinations of petroleum geoscience topic associations, by:
creation of a supervised machine learning statistical model using contradiction scores and their corresponding historical business success/failure, through such techniques as (but not limited to) Naive Bayes, Random Forest, Support Vector Machines and Neural Networks.
