US Patent Application 18324315. ARTIFICIAL INTELLIGENCE ENGINE FOR ENTITY RESOLUTION AND STANDARDIZATION simplified abstract

ARTIFICIAL INTELLIGENCE ENGINE FOR ENTITY RESOLUTION AND STANDARDIZATION

Organization Name

Mastercard International Incorporated

Inventor(s)

Yogesh Sakpal of Mumbai (IN)

Gauri Shah Bhatnagar of Meerut (IN)

Shraddha Shirke of Kalwa (IN)

Dean Vaz of Hyderabad (IN)

Siddhesh Dongare of Kalyan (IN)

Dmitriy Kontarev of Allen TX (US)

Brett Ragozzine of Lehi UT (US)

Christopher Brousseau of Provo UT (US)

ARTIFICIAL INTELLIGENCE ENGINE FOR ENTITY RESOLUTION AND STANDARDIZATION - A simplified explanation of the abstract

This abstract first appeared for US patent application 18324315 titled 'ARTIFICIAL INTELLIGENCE ENGINE FOR ENTITY RESOLUTION AND STANDARDIZATION

Simplified Explanation

- The patent application describes techniques for training an entity resolution model. - The entity resolution model is trained using raw training data, which includes historical transaction data. - A label dictionary is generated by performing natural language processing (NLP) on the training data. - The label dictionary includes extracted entities from the text of each transaction. - Tagged data is generated from the training data using the label dictionary. - Vocabulary training is performed on the training data by tokenizing the text of each transaction and converting it into a transformer model specific format. - The entity resolution model is trained using a transformer model, which utilizes the tokenized text and the tagged data. - The trained entity resolution model is stored in a database for future use.

Original Abstract Submitted

Techniques for training an entity resolution model are presented. The techniques include inputting raw training data into the entity resolution model. The training data includes historical transaction data including a plurality of transactions. A label dictionary is generated by performing natural language processing (NLP) on the training data. The NLP includes scanning text of each transaction, extracting one or more entities from the text, and storing the label dictionary in a database. The label dictionary includes the extracted entities. Tagged data is generated from the training data using the label dictionary. Vocabulary training is performed on the training data, including tokenizing the text of each transaction and converting the tokenized text into a transformer model specific format. The entity resolution model is then trained using a transformer model, which uses the tokenized text and the tagged data. The trained entity resolution model is then stored in a database.