Yandex has filed a patent for a method and server to factually validate content in snippets. By using Natural Language Processing (NLP) models, the system determines the accuracy of a given snippet by analyzing the context provided by other snippets. This technology aims to improve the accuracy of search results. GlobalData’s report on Yandex gives a 360-degree view of the company including its patenting strategy. Buy the report here.
Smarter leaders trust GlobalData
According to GlobalData’s company profile on Yandex, Social media analytics was a key innovation area identified from patents. Yandex's grant share as of January 2024 was 46%. Grant share is based on the ratio of number of grants to total number of patents.
Factually validating content of snippets using nlp model
The patent application (Publication Number: US20240012838A1) describes a computer-implemented method for factually validating the content of a given snippet among a plurality of snippets generated by a generative machine-learning model in response to a search query. The method involves identifying another snippet providing context to the given snippet, generating target and context vectors using a text embedding algorithm, and feeding them to a Natural Language Processing (NLP) model to determine the probability of factual accuracy confirmation. The NLP model is trained to assess the coherence and factual accuracy of snippets, enhancing the validation process.
Furthermore, the method incorporates the use of a Transformer-based machine-learning model to generate cohesive text snippets for the search object. The NLP model is fine-tuned with an add-on machine learning model, utilizing training objects with prompts, contextually related snippets, and labels for factual confirmation. The approach involves freezing weights of the NLP model after pre-training and applying a P-tuning approach for fine-tuning. The server executing the method is equipped with a processor and memory storing instructions to identify contextually related snippets, generate vectors, and assess factual accuracy probabilities. The server's configuration allows for efficient validation of snippet content based on semantic coherence and factual confirmation, enhancing the accuracy of search results.
To know more about GlobalData’s detailed insights on Yandex, buy the report here.
Premium 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.
