Microsoft Technology Licensing, LLC

Informatica Corporation

Manish Gupta

Prashant Gupta

Joy Mustafi

Gregorio Convertino

Puneet Agrawal

Mithun Das Gupta

Chitresh Kakwani

Ranjeet Kumar Tayi

Swati Tomar

Manoj Kumar Agarwal

Kishor Chamua

Srimukha Paturi

Kumar Subham

Ambuj Pushkar Ojha

Rishab

Deepanjali Singh

Anuska Nandy

Tezan Sahu

Akriti

Systems and methods for generating autocomplete text using a language model are disclosed. An image and text-prefix may be entered at an input field of a search application. The image is processed to generate an image description. The image description and the text-prefix signals may be used as input at a language model to generate an autocomplete text by the language model. A contextual history may also be included as input to the language model. The autocomplete text is an output by the language model based on the input at the language model. The auto-complete text may be a next-word ghosting.

Ghosting for multimodal dialogs

This disclosure describes a query gateway system that provides an efficient and flexible framework for providing context-retained autosuggest queries from an autosuggest query system (e.g., a search engine query experience) to a generative language model system (e.g., an AI chat experience). For instance, the query gateway system establishes a framework to leverage the features and services of the autosuggest query system and automatically provides context-retained queries to the generative language model system using separate user interfaces that do not disrupt user navigation or require manual duplicative user input. Additionally, the query gateway system incorporates additional enhancements, including an AI chat eligibility model and a query reformulation model, to improve the computational efficiency and accuracy of the AI chat system.

Generating narrative query responses utilizing generative language models from search-based autosuggest queries

Aspects of the present disclosure relate to providing, based on a partial query string, a plurality of autosuggestions that are diverse in nature such that the user is more likely to see the preferred complete query terms and therefore more likely to select one of the preferred suggestions which will increase search efficiency. As described herein, such functionality relates to generating cluster groups of candidate suggestions, each cluster including sub-topics, then performing the search based on a selected cluster or sub-topic. To generate the cluster groups, systems and methods, as described herein, analyze the similarity between candidate suggestions as well as the popularity of generated sub-topics. The cluster groups and sub-topics may be displayed visually, and in certain embodiments the cluster groups are ordered vertically from top to bottom and aligned to the left side of the display, while sub-topics are ordered horizontally from left to right following the cluster label.

Clustering suggestions for partial query auto-completion

A system, method and computer-readable medium for missing data identification, including identifying columns in tables of a database, generating categorical columns of categorical data by transforming data values in the columns into categorical data values, generating a co-occurrence matrix corresponding to a pair of categorical columns in the categorical columns, determining an expected frequency of co-occurrence corresponding to each unique pair of categorical data values based at least in part on a marginal totals corresponding to categorical data values in the co-occurrence matrix, and identifying one or more locations of missing data based at least in part on the count of co-occurrence of each unique pair of categorical data values and the expected frequency of co-occurrence corresponding to each unique pair of categorical data values.

Method, apparatus, and computer-readable medium for missing data identification

A system, method and computer-readable medium for generation of missing data including transmitting indicators corresponding to locations of missing data in columns in tables of a database, each location of missing data corresponding to categorical values of categorical columns and each location of missing data being identified based on an expected count of data values at the corresponding location, receiving a selection of at least one indicator corresponding to at least one location of missing data, the at least one location of missing data corresponding to two or more categorical values of two or more categorical columns in the categorical columns, and generating sets of data records in at least one table in the tables of the database, each set of data records having two or more column values in two or more columns that correspond to the two or more categorical values of the two or more categorical columns.

Method, apparatus, and computer-readable medium for generation of missing data

Techniques for automating the generation and analysis of fundus drawings are described. Captured images undergo image processing to extract information about image features. Fundus images are generated and recommended labels for the fundus drawing are generated. Fundus drawings can be analyzed and undergo textual processing to extract existing labels. Machine learning models and co-occurrence analysis can be applied to collections of fundus images and drawings to gather information about commonly associated labels, label locations, and user information. The most frequently used labels associated with the image can be identified to improve recommendations and personalize labels.

Automatic generation of fundus drawings

Aspects extend to methods, systems, and computer program products for forecasting eye condition progression for eye patients. When a patient visits an eye practitioner, the patient (or when appropriate their guardian) may be interested in the current eye condition as well as a prediction of eye condition progression in the future and/or as the patient ages. Aspects of the invention can be used to predict the progress of an eye condition for a patient (e.g., a child) at a number of different post-examination times after an examination. Predicting the progress of an eye condition for a patient over time can be used to assist the eye practitioner in tailoring a treatment plan and/or tailoring a subsequent examination schedule for the patient.

Forecasting eye condition progression for eye patients

Aspects extend to methods, systems, and computer program products for providing personalized surgery recommendations for eye patients. Surgery types, and surgery parameters can be recommended for a patient based on predicted post-operative UCVA for the patient if the surgery types and surgery parameters were to be used. Predicting post-operative UCVA can be handled as a regression problem based on patient demography and pre-operative examination details. In an additional aspect, surgery parameters are automatically determined and/or optimized for improved post-operative UCVA by including surgery parameters in a regression model.

Personalized refractive surgery recommendations for eye patients

Disclosed in some examples are methods, systems, devices, and machine readable mediums that automatically determine a set of one or more rules for finding items of content with high precision and high-recall. The disclosed methods enable applications to find content in rapidly changing datasets such as microblogs and other social networking feeds with a high precision and high recall, improving the accuracy of the application.

Growing community of inventors

Hyderabad Telangana, India

Manish Gupta