利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

10 条记录 (耗时 0.013 秒)

    SINGLE MODEL-BASED BEHAVIOR PREDICTIONS IN AN ON-DEMAND ENVIRONMENT
    2.
    发明申请
    SINGLE MODEL-BASED BEHAVIOR PREDICTIONS IN AN ON-DEMAND ENVIRONMENT 审中-公开

    公开(公告)号：US20180096267A1

    公开(公告)日：2018-04-05

    申请号：US15712911

    申请日：2017-09-22

    申请人： salesforce.com, inc.

    发明人： Chalenge Masekera ,  Vitaly Gordon ,  Leah McGuire ,  Shubha Nabar

    IPC分类号： G06Q10/06 ,  G06N5/04 ,  G06N99/00 ,  G06Q10/08 ,  G06Q10/04 ,  G06Q30/02

    CPC分类号： G06Q10/06 ,  G06F16/00 ,  G06N5/04 ,  G06N20/00 ,  G06Q10/04 ,  G06Q10/08 ,  G06Q30/0201

    摘要： In accordance with embodiments, there are provided mechanisms and methods for facilitating single model-based behavior predictions in an on-demand services environment in an on-demand services environment according to one embodiment. In one embodiment and by way of example, a method comprises collecting, by a model selection and application server device (“model device”), information associated with customers of a tenant, and extracting, from the information, behavior traits of the customers as they relate to products or services offered by the tenant. The method further includes dynamically selecting, by the model device, a single model from a set of models to convert the behavior traits into predictions indicating anticipated conduct of each customer in relation to one or more products or one or more of the services of the tenant, where the single model performs multiple processes to convert the behavior traits into predictions, and where the multiple processes include at least two of the following: evaluating data, cleansing the data, transforming the data. The method may further include writing the data, and transmitting, over a communication medium, the predictions to the tenant.

    REDUCING INSTANCES OF INCLUSION OF DATA ASSOCIATED WITH HINDSIGHT BIAS IN A TRAINING SET OF DATA FOR A MACHINE LEARNING SYSTEM
    3.
    发明申请
    REDUCING INSTANCES OF INCLUSION OF DATA ASSOCIATED WITH HINDSIGHT BIAS IN A TRAINING SET OF DATA FOR A MACHINE LEARNING SYSTEM 审中-公开

    公开(公告)号：US20200057959A1

    公开(公告)日：2020-02-20

    申请号：US16264659

    申请日：2019-01-31

    申请人： salesforce.com, inc.

    发明人： Kevin Moore ,  Leah McGuire ,  Matvey Tovbin ,  Mayukh Bhaowal ,  Shubha Nabar

    IPC分类号： G06N20/00

    摘要： Instances of data associated with hindsight bias in a training set of data for a machine learning system can be reduced. A first set of data, having a first set of fields, can be received. Data in a first field can be analyzed with respect to data in a second field corresponding to an event to be predicted. A result can be that the data in the first field is associated with hindsight bias. A second set of data, having a second set of fields, can be produced. The second set of fields can exclude the first field. One or more features associated with the second set of data can be generated. A third set of data, having the second set of fields and fields that correspond to the one or more features, can be produced. The training set of data can be produced using the third set of data.

    Identification and application of hyperparameters for machine learning
    4.
    发明授权
    Identification and application of hyperparameters for machine learning 有权

    公开(公告)号：US11526799B2

    公开(公告)日：2022-12-13

    申请号：US16264583

    申请日：2019-01-31

    申请人： salesforce.com, inc.

    发明人： Kevin Moore ,  Leah McGuire ,  Eric Wayman ,  Shubha Nabar ,  Vitaly Gordon ,  Sarah Aerni

    IPC分类号： G06N20/00 ,  G06N5/04

    摘要： Methods and systems are provided to determine suitable hyperparameters for a machine learning model and/or feature engineering process. A suitable machine learning model and associated hyperparameters are determined by analyzing a dataset. Suitable hyperparameter values for compatible machine learning models having one or more hyperparameters in common and a compatible dataset schema are identified. Hyperparameters may be ranked according to each of their respective influences on a model performance metrics, and hyperparameter values identified as having greater influence may be more aggressively searched.

    IDENTIFICATION AND APPLICATION OF HYPERPARAMETERS FOR MACHINE LEARNING
    6.
    发明申请
    IDENTIFICATION AND APPLICATION OF HYPERPARAMETERS FOR MACHINE LEARNING 审中-公开

    公开(公告)号：US20200057958A1

    公开(公告)日：2020-02-20

    申请号：US16264583

    申请日：2019-01-31

    申请人： salesforce.com, inc.

    发明人： Kevin Moore ,  Leah McGuire ,  Eric Wayman ,  Shubha Nabar ,  Vitaly Gordon ,  Sarah Aerni

    IPC分类号： G06N20/00 ,  G06N5/04

    摘要： Methods and systems are provided to determine suitable hyperparameters for a machine learning model and/or feature engineering process. A suitable machine learning model and associated hyperparameters are determined by analyzing a dataset. Suitable hyperparameter values for compatible machine learning models having one or more hyperparameters in common and a compatible dataset schema are identified. Hyperparameters may be ranked according to each of their respective influences on a model performance metrics, and hyperparameter values identified as having greater influence may be more aggressively searched.

    Recognition of biases in data and models
    7.
    发明授权
    Recognition of biases in data and models 有权

    公开(公告)号：US10984283B2

    公开(公告)日：2021-04-20

    申请号：US16565922

    申请日：2019-09-10

    申请人： salesforce.com, inc.

    发明人： Sarah Aerni ,  Natalie Casey ,  Shubha Nabar ,  Melissa Runfeldt ,  Sara Beth Asher

    IPC分类号： G06K9/62 ,  G06F17/15

    摘要： A method of training a predictive model to predict a likely field value for one or more user selected fields within an application. The method comprises providing a user interface for user selection of the one or more user selected fields within the application; analyzing a pre-existing, user provided data set of objects; training, based on the analysis, the predictive model; determining, for each user selected field based on the analysis, a confidence function for the predictive model that identifies the percentage of cases predicted correctly at different applied confidence levels, the percentage of cases predicted incorrectly at different applied confidence levels, and the percentage of cases in which the prediction model could not provide a prediction at different applied confidence levels; and providing a user interface for user review of the confidence functions for user selection of confidence threshold levels to be used with the predictive model.

    RECOGNITION OF BIASES IN DATA AND MODELS
    8.
    发明申请
    RECOGNITION OF BIASES IN DATA AND MODELS 有权

    公开(公告)号：US20210073579A1

    公开(公告)日：2021-03-11

    申请号：US16565922

    申请日：2019-09-10

    申请人： salesforce.com, inc.

    发明人： Sarah Aerni ,  Natalie Casey ,  Shubha Nabar ,  Melissa Runfeldt ,  Sara Beth Asher

    IPC分类号： G06K9/62 ,  G06F17/15

    摘要： A method of training a predictive model to predict a likely field value for one or more user selected fields within an application. The method comprises providing a user interface for user selection of the one or more user selected fields within the application; analyzing a pre-existing, user provided data set of objects; training, based on the analysis, the predictive model; determining, for each user selected field based on the analysis, a confidence function for the predictive model that identifies the percentage of cases predicted correctly at different applied confidence levels, the percentage of cases predicted incorrectly at different applied confidence levels, and the percentage of cases in which the prediction model could not provide a prediction at different applied confidence levels; and providing a user interface for user review of the confidence functions for user selection of confidence threshold levels to be used with the predictive model.