中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

6 records (took 0.015 seconds)

    FEATURE GENERATION FOR ASSET CLASSIFICATION
    2.
    发明申请
    FEATURE GENERATION FOR ASSET CLASSIFICATION 有权

    公开(公告)号：US20210248457A1

    公开(公告)日：2021-08-12

    申请号：US16785191

    申请日：2020-02-07

    Applicant: International Business Machines Corporation

    Inventor： Omar Odibat ,  Jean Xu Yu ,  Emrah Zarifoglu ,  Ilyas Mohamed Iyoob

    IPC: G06N3/08 ,  G06N3/04 ,  H04L29/08

    Abstract: An embodiment includes generating an input document that includes a plurality of text fields of attribute data. The embodiment also includes extracting a set of candidate features from the attribute data using a feature extraction module that evaluates the attribute data using first and second machine learning models, where the first machine learning model scores terms in the input document and the second machine learning model includes a deep learning model. The embodiment also includes calculating feature-selection values for respective features of the set of candidate features and generating tag information for a remote computing asset using a machine learning classifier that predicts the tag information based on the feature-selection values.

    CONSTRAINED LARGE-DATA MARKDOWN OPTIMIZATIONS BASED UPON MARKDOWN BUDGET
    3.
    发明申请
    CONSTRAINED LARGE-DATA MARKDOWN OPTIMIZATIONS BASED UPON MARKDOWN BUDGET 审中-公开
    基于标记预算的约束数据标记优化

    公开(公告)号：US20170032415A1

    公开(公告)日：2017-02-02

    申请号：US14814743

    申请日：2015-07-31

    Applicant: International Business Machines Corporation

    Inventor： Jun Lei Chen ,  Xiao Chun Li ,  Emily A. Port ,  Siqun Wang ,  Emrah Zarifoglu

    IPC: G06Q30/02

    CPC classification number: G06Q30/0249

    Abstract: A markdown budget user interface is provided that implements access to large-scale computational resources capable of concurrently manipulating a number of multi-million item data sets and that allows a retailer to specify inputs. The inputs include a markdown budget constraint that applies across all retailer stores, a group of store-product data sets that each include initial prices and markdown start prices of different combinations of products within all of the retailer stores, and a markdown objective, selected from a group including profit, revenue, and sales volume, to be maximized within the markdown budget constraint. The inputs are received from the retailer using the markdown budget user interface and the large-scale computational resources are invoked to determine a markdown recommendation that includes an indication of the store-product data sets that satisfy both the markdown budget constraint and the markdown objective. The markdown recommendation is provided to the retailer.

    Abstract translation: 提供了降价预算用户界面，其实现对能够同时操作数百万个项目数据集的大规模计算资源的访问，并且允许零售商指定输入。 输入包括适用于所有零售商商店的降价预算约束，一组商店产品数据集，每个商品数据集包括所有零售商商店内不同产品组合的初始价格和降价开始价格，以及从 包括利润，收入和销售量在内的集团将在降价预算约束内最大化。 使用降价预算用户界面从零售商接收输入，并且调用大规模的计算资源来确定包括满足降价预算约束和降价目标的商品数据集的指示的降价推荐。 向零售商提供降价建议。

    Avoidance of product stockouts through optimized routing of online orders
    4.
    发明授权
    Avoidance of product stockouts through optimized routing of online orders 有权

    公开(公告)号：US10956859B2

    公开(公告)日：2021-03-23

    申请号：US15995272

    申请日：2018-06-01

    Applicant: International Business Machines Corporation

    Inventor： France Savard ,  Xiaowei Bao ,  Emrah Zarifoglu ,  Pavel Zelinsky

    IPC: G06Q10/08 ,  G06Q30/02 ,  G06Q30/06

    Abstract: A method, system and computer program product for fulfilling an online order. An online order to purchase an item(s) is received. The “candidate locations” that stock the item(s) of the online order and that can be used to fulfill at least a portion of the online order are determined. A stockout cost for each of these candidate locations for fulfilling an item of the online order may be calculated, where the stockout cost is a cost of a potential lost sale of the item of the online order by the candidate location if the candidate location fulfills the item of the online order. A shipping location among the candidate locations to fulfill the item is then determined based at least in part on the stockout cost for each of the candidate locations for fulfilling the item. The item is then shipped to the customer from the determined shipping location.

    Modeling customer demand and updating pricing using customer behavior data
    5.
    发明授权
    Modeling customer demand and updating pricing using customer behavior data 有权

    公开(公告)号：US10832268B2

    公开(公告)日：2020-11-10

    申请号：US15409806

    申请日：2017-01-19

    Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

    Inventor： Norbert M. Binkiewicz ,  Junlei Chen ,  Elizabeth J. Chester ,  Prafulla N. Dawadi ,  Robert K. Parkin ,  Emrah Zarifoglu

    IPC: G06Q10/00 ,  G06Q30/02 ,  G06Q30/06 ,  H04L29/08

    Abstract: In an aspect of the invention, a computer-implemented method includes: receiving, by a computing device via computer network, a plurality of session data records indicating computer network browsing activity between a plurality of client devices and a merchant server hosting an online store; aggregating, by the computing device, a subset of the plurality of session data records for a single product, of a plurality of products, identified in the session data records and offered for purchase by the online store; extracting, by the computing device, features from the aggregated subset of session data records relating to customer demand for a the single product; modeling, by the computing device, customer demand for the single product based on the extracted features; optimizing, by the computing device, a price for the single product based on results of the modeling; and publishing, by the computing device, the optimized price.

Patent Agency Ranking