知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

18 records (took 0.017 seconds)

    CRASH LOCALIZATION USING CRASH FRAME SEQUENCE LABELLING
    5.
    发明申请
    CRASH LOCALIZATION USING CRASH FRAME SEQUENCE LABELLING 有权

    公开(公告)号：US20230091899A1

    公开(公告)日：2023-03-23

    申请号：US17483571

    申请日：2021-09-23

    Applicant: Microsoft Technology Licensing, LLC

    Inventor： Chetan BANSAL ,  Manish Shetty MOLAHALLI ,  Suman Kumar NATH ,  Siamak AHARI ,  Haitao WANG ,  Sean A. BOWLES ,  Kamil Ozgur ARMAN

    IPC: G06F11/36

    Abstract: Machine-learned prediction of a blame frame of a crash stack. Specifically, a crash stack associated with a crash is parsed into a sequence of frames. The blame frame of the crash stack is estimated by, for each of a plurality of the sequence of frames, identifying a plurality of features of the corresponding frame, feeding the plurality of features to a neural network, and using the output of the neural network to make a prediction on whether the corresponding frame is a blame frame of the crash. If this is done during training time, the predicted blame frame can be compared against the actual blame frame, resulting in an adjustment of the neural network. Through appropriate featurization of the frames, and by use of the neural network, the prediction can be made cross-application and considering the context of the frame within the crash stack.

    Providing and Leveraging Implicit Signals Reflecting User-to-BOT Interaction
    6.
    发明申请
    Providing and Leveraging Implicit Signals Reflecting User-to-BOT Interaction 审中-公开

    公开(公告)号：US20190149488A1

    公开(公告)日：2019-05-16

    申请号：US15810049

    申请日：2017-11-11

    Applicant: Microsoft Technology Licensing, LLC

    Inventor： Chetan BANSAL ,  Anantha Deepthi UPPALA

    IPC: H04L12/58 ,  G06F9/44 ,  H04L29/08

    CPC classification number: H04L51/02 ,  G06F9/453 ,  H04L51/14 ,  H04L67/22 ,  H04L67/32

    Abstract: A technique is described herein for providing implicit quality signals over a span of time that reflect quality of service provided by a collection of BOTs to a group of users. The technique can then leverage these implicit quality signals in various application-phase uses. In one use, an abandonment-determination component can leverage the implicit quality signals to provide an output result which indicates whether a current user has abandoned use of a current BOT with which he or she has been interacting, or is about to abandon use of that current BOT. In another use, a search engine or a recommendation engine can use the implicit quality signals to help identify an appropriate BOT for use by the current user. The implicit quality signals can include: one of more user-behavior implicit signals; one of more BOT-behavior implicit signals; and/or one of more transaction-summary implicit signals.

    AUTOMATED TROUBLESHOOTER
    9.
    发明申请
    AUTOMATED TROUBLESHOOTER 有权

    公开(公告)号：US20220414463A1

    公开(公告)日：2022-12-29

    申请号：US17743067

    申请日：2022-05-12

    Applicant: MICROSOFT TECHNOLOGY LICENSING, LLC

    Inventor： Rahul MITTAL ,  Manish Shetty MOLAHALLI ,  Puneet KAPOOR ,  Chetan BANSAL ,  Tarun SHARMA ,  Abhilekh MALHOTRA ,  Sunil SINGHAL

    IPC: G06N3/08 ,  G06N5/02

    Abstract: The technology described herein generates automated workflows from trouble shooting guides. The automated workflow generation process described herein starts with existing TSGs as the input. A first step in the process may be identifying the computer commands in the TSG. In one aspect, the commands are identified using a sequence-to-sequence model. Once a command is identified as a command, the command is associated with an application of origin. In aspects, a second model is used to identify the application associated with the command. The second model may be a metric-based meta-learning approach to associate a command with an application. Once the commands are identified and associated with an application, they may be parsed or extracted using a regular expression, which is a special text string describing a search pattern. The structure of the natural text is then parsed to build an executable decision tree and merged with the parsed commands.

    EXTRAQUERY CONTEXT-AIDED SEARCH INTENT DETECTION
    10.
    发明申请
    EXTRAQUERY CONTEXT-AIDED SEARCH INTENT DETECTION 有权

    公开(公告)号：US20220107802A1

    公开(公告)日：2022-04-07

    申请号：US17062581

    申请日：2020-10-03

    Applicant: Microsoft Technology Licensing, LLC

    Inventor： Nikitha RAO ,  Chetan BANSAL ,  Zhongyan GUAN ,  Mark Alistair WILSON-THOMAS ,  Nachiappan NAGAPPAN ,  Thomas Michael Josef ZIMMERMANN

    IPC: G06F8/77 ,  G06F16/953 ,  G06F16/2453 ,  G06N20/00

    Abstract: Embodiments promote searcher productivity and efficient search engine usage by using extraquery context to detect a searcher's intent, and using detected intent to match searches to well-suited search providers. Extraquery context may include cursor location, open files, and other editing information, tool state, tool configuration or environment, project metadata, and other information external to actual search query text. Search intent may be code (seeking snippets) or non-code (seeking documentation), and sub-intents may be distinguished for different kinds of documentation or different programming languages. Search provider capabilities may reflect input formats such as natural language or logical operator usage, or content scope such as web-wide or local, or other search provider technical characteristics. Search intent detection permits efficient and effective use of a single search box for a wide variety of different searches for different kinds of results, thereby simplifying a development tool user interface.

Patent Agency Ranking