Theory-Guided Lithostratigraphic Analogue Mining From Unstructured Text
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional information retrieval systems fail to accurately identify geological analogues in geoscience texts due to biases from local geographical and temporal factors, leading to inaccurate similarity computations and missed valuable insights.
Innovation Solution
A theory-guided data science method that computes geological Lithostratigraphic analogues by filtering out geographical entities, applying a hierarchical entity association, and weighting nouns, adjectives, and verbs in text vectors to prioritize meaningful similarities, with an entropy-based reliability score and user feedback mechanism.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional information retrieval systems are used to identify geological analogues, then the system is simple to operate, but the accuracy of analogue identification is low due to biases from local geographical and temporal factors
Solution Approach 1:
The patent extracts and removes geographical entities and temporal information from the text data before computing similarity. This extraction eliminates the bias introduced by local geographical and temporal factors, allowing the system to identify genuine geological analogues based on lithostratigraphic characteristics rather than location-specific context.
Solution Approach 2:
The patent changes the parameter weighting in similarity computation by assigning higher weights to nouns, adjectives, and verbs that describe geological characteristics, while giving lower or zero weight to geographical and temporal parameters. This parameter adjustment prioritizes geologically meaningful similarities over geographical proximity.
2Loss of information
If all text entities are weighted equally in similarity computation, then the computation is simple, but valuable insights are missed due to lack of prioritization
Solution Approach 1:
The patent applies local quality by differentiating the importance of different text entities based on their grammatical category and geological relevance. Nouns, adjectives, and verbs that describe lithostratigraphic characteristics are assigned higher weights, while other entities receive lower weights, creating a non-uniform weighting scheme that prioritizes geologically significant information.
Solution Approach 2:
The patent creates weighted versions of text vectors where each entity is replicated with a weight coefficient reflecting its importance. This allows the original text data to be preserved while generating enhanced representations that emphasize critical geological features for more accurate analogue identification.
3Reliability
If geographical entities are included in text vectors, then the text representation is complete, but bias from local factors contaminates similarity scores
Solution Approach 1:
The patent extracts geographical entities from the text and removes them from the vector representation used for similarity computation. This extraction preserves the information for separate analysis while eliminating its contaminating effect on geological similarity scores, allowing reliable identification of analogues based on geological characteristics alone.
Data Source
AI summary
The invention is a data processing method and system for assisting geoscientists in identifying geological Lithostratigraphic analogues from unstructured text using theory-guided machine learning. The data processing system makes the necessary calculations to detect geological Lithostratigraphic entities and appropriate entity relations, accounting for space (local geography) and time (local succession) when using word associations to determine similarity. The system computes the reliability of that score, incorporating user feedback into the learned model. In particular, the data processing system operates on any digital unstructured text derived from academic literature, company reports, web pages and other sources. Similarity between the associations of words to geological Lithostratigraphic entities can be used to suggest global analogues to communities such as petroleum geoscientists in order to reduce petroleum exploration risk and improve field development decisions.