Automated Metadata Extraction for Wellbore Planning Documents
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for identifying subsurface exploration documents in a database are manual and time-consuming, leading to inefficiencies and errors in document identification for wellbore planning.
Innovation Solution
A method and system utilizing machine-learned models and natural language processing to automatically determine document categories, types, and metadata attributes, enabling efficient retrieval and planning of wellbores using earth property data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual labelling of exploration documents with metadata is used, then document identification capability is achieved, but time consumption and error rate increase
Solution Approach 1:
The system enables documents to automatically generate their own metadata through machine-learned models that process document content, titles, and earth property data. The NLP algorithm extracts metadata attributes autonomously without requiring manual human intervention, making the documentation process self-servicing.
Solution Approach 2:
The manual mechanical process of labelling documents is replaced with an automated electronic system using machine-learned models and natural language processing algorithms. These computational systems automatically analyze document content and generate metadata, substituting human manual work with algorithmic processing.
2Productivity
If manual labelling of exploration documents is used, then document categorization is achieved, but error rate increases
Solution Approach 1:
The machine-learned models are trained using feedback from labeled training data, continuously improving their accuracy. The system processes documents through multiple stages including preprocessing, model inference, and validation, with the ability to learn from correct and incorrect classifications to enhance future performance.
Solution Approach 2:
Human manual labelling is replaced with automated machine-learned models and NLP algorithms that consistently apply classification rules without human error. The electronic processing system maintains uniform standards across all documents, eliminating variability inherent in manual human work.
3Productivity
If automated machine-learned models are used for metadata extraction, then processing efficiency is improved, but system complexity increases
Solution Approach 1:
The automated system is divided into distinct modular components: a preprocessing module that prepares documents, machine-learned models that extract specific metadata attributes, and an NLP algorithm that processes text content. Each module performs a specific function and can be independently trained, maintained, and improved.
Solution Approach 2:
The machine-learned models and NLP algorithm serve multiple functions: they process different document types (seismic surveys, electromagnetic surveys, well log data), extract various metadata attributes (category, type, properties), and work with different data formats. This multi-functionality reduces the need for separate specialized systems.
Data Source
AI summary
A method includes obtaining a document comprising earth property data regarding a geological region of interest, preprocessing the document to form at least one preprocessed document and determining, using a set of trained machine-learned models processing the at least one preprocessed document, a category of the document and a type of the document. The method further includes determining, using a natural language processing algorithm, metadata attributes of the document and a title of the document, and updating a database storing the document with the title, the category, the type and the metadata attributes. The method further includes identifying, by a planning module processing a query, the document from the database based on at least one of the title, the category, the type and the metadata attributes and planning a wellbore path in the geological region of interest using the earth property data comprised in the document.