Salesforce, inc. (20240338961). PROCESSING FORMS USING ARTIFICIAL INTELLIGENCE MODELS simplified abstract
Contents
PROCESSING FORMS USING ARTIFICIAL INTELLIGENCE MODELS
Organization Name
Inventor(s)
Mingfei Gao of Sunnyvale CA (US)
Ran Xu of Mountain View CA (US)
PROCESSING FORMS USING ARTIFICIAL INTELLIGENCE MODELS - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240338961 titled 'PROCESSING FORMS USING ARTIFICIAL INTELLIGENCE MODELS
The abstract describes a patent application for a technology that allows an application server to extract information from input documents using optical character recognition and machine learning models to identify key values based on input queries.
- The application server receives an input document with text fields and a key phrase querying a value for a key-value pair.
- It extracts character strings and their locations on the document using optical character recognition.
- The extracted information is input into a machine learned model to compute the probability that a character string corresponds to the value for the key-value pair.
- The server then identifies the value for the key-value pair based on the input key phrase and outputs the identified value.
Potential Applications: - Document processing and data extraction - Automated form filling and data retrieval - Information retrieval in various industries such as finance, healthcare, and legal
Problems Solved: - Streamlining data extraction processes - Improving accuracy and efficiency in identifying key values - Enhancing automation in document processing tasks
Benefits: - Time-saving in data extraction tasks - Increased accuracy in identifying key values - Reduction in manual data entry errors
Commercial Applications: Title: Automated Data Extraction Technology for Document Processing This technology can be used in industries such as finance, insurance, and healthcare for automating data extraction processes, improving efficiency, and reducing manual errors.
Questions about the technology: 1. How does this technology improve the accuracy of data extraction from documents? - The technology uses machine learning models to compute the probability that a character string corresponds to the value for the key-value pair, enhancing accuracy. 2. What are the potential challenges in implementing this technology in different industries? - Some challenges may include adapting the technology to different document layouts and formats, as well as ensuring compatibility with existing systems.
Original Abstract Submitted
an application server may receive an input document including a set of input text fields and an input key phrase querying a value for a key-value pair that corresponds to one or more of the set of input text fields. the application server may extract, using an optical character recognition model, a set of character strings and a set of two-dimensional locations of the set of character strings on a layout of the input document. after extraction, the application server may input the extracted set of character strings and the set of two-dimensional locations into a machine learned model that is trained to compute a probability that a character string corresponds to the value for the key-value pair. the application server may then identify the value for the key-value pair corresponding to the input key phrase and may out the identified value.
