Snowflake had nine patents in artificial intelligence during Q2 2024. Snowflake Inc’s patents in Q2 2024 focus on streamlining Natural Language Processing (NLP) model generation through iterative training, enhancing questionnaire completion systems on a data platform, and optimizing computing resource allocation for query execution based on metadata and machine learning predictions. These technologies aim to improve efficiency, accuracy, and security in data retrieval and processing within automated systems. GlobalData’s report on Snowflake gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
Snowflake had no grants in artificial intelligence as a theme in Q2 2024.
Recent Patents
Application: Iterative training for multi-modal data in natural language processing (Patent ID: US20240211691A1)
The patent filed by Snowflake Inc. describes systems and methods for generating a Natural Language Processing (NLP) model through iterative training. The method involves processing real-world documents containing text, layout, and image data using a neural network to generate initial predictions, which are then validated and corrected based on the information present in the documents. The NLP model's quality is evaluated, and once a quality constraint is met, the model is configured to extract data points from new documents without further validation. This approach enhances the efficiency and accuracy of data retrieval in automated systems by streamlining information extraction from diverse document formats.
The system and method outlined in the patent involve multiple iterations to train the NLP model, with each iteration including processing real-world documents, validating predictions, and configuring the model for new document processing. The system can train the model using validated predictions as training data, stop iterations once quality constraints are met, and generate outputs based on text, layout, and image data. Additionally, user validation of the model's output is incorporated, allowing users to confirm, correct, or mark extracted data points. The patent also describes the use of layout-aware and multi-modal models within an encoder-decoder framework, adjusting models based on user feedback, and continuing iterations until a predetermined accuracy threshold is achieved. The NLP model can be configured to extract key information, classify documents, and answer natural language questions, demonstrating a comprehensive and adaptable approach to NLP model generation.
To know more about GlobalData’s detailed insights on Snowflake, buy the report here.
Data Insights
From
The gold standard of business intelligence.
Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.
GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.
GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.
