公开(公告)号：US20170046779A1

公开(公告)日：2017-02-16

申请号：US14826431

申请日：2015-08-14

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Feng Cao ,  Heng Cao ,  Yuan Ni ,  Fei Su

IPC: G06Q40/02

CPC classification number: G06Q40/025 ,  G06N7/005 ,  G06N7/08

Abstract: A method, a computer program product, and a computer system for making a decision of long step and healthy credit limit enhancement. A computer constructs a Markov decision process graph which includes nodes and edges, wherein the nodes represent respective states of one or more customer segments and the edges represent paths based on historical data. The computer applies long step actions for a respective one of the one or more customer segments, wherein the long step actions enhance more than one credit limit levels. The computer calculates gained values of the long step actions. The computer chooses an optimal long step action from the long step actions, wherein the optimal long step action has a maximum gained value.

Abstract translation: 一种方法，计算机程序产品和计算机系统，用于做出长期健康的信用限度增长的决定。 计算机构建包括节点和边缘的马尔可夫决策过程图，其中节点表示一个或多个客户分段的相应状态，并且边缘表示基于历史数据的路径。 计算机对一个或多个客户分段中的相应一个客户分段应用长步骤动作，其中长步骤动作增强多于一个信用限额水平。 计算机计算长步骤动作的获得值。 计算机从长步骤动作中选择最佳的长步骤动作，其中最佳长步骤动作具有最大获得值。

公开(公告)号：US20170046780A1

公开(公告)日：2017-02-16

申请号：US14980570

申请日：2015-12-28

Applicant: International Business Machines Corporation

Inventor： Feng Cao ,  Heng Cao ,  Yuan Ni ,  Fei Su ,  Hui Jia Zhu

IPC: G06Q40/02 ,  G06N7/08

CPC classification number: G06Q40/025 ,  G06N7/005 ,  G06N7/08

Abstract: A method for making a decision of long step and healthy credit limit enhancement. A computer constructs a Markov decision process graph which includes nodes and edges, wherein the nodes represent respective states of one or more customer segments and the edges represent paths based on historical data. The computer applies long step actions for a respective one of the one or more customer segments, wherein the long step actions enhance more than one credit limit levels. The computer calculates gained values of the long step actions. The computer chooses an optimal long step action from the long step actions, wherein the optimal long step action has a maximum gained value.

Abstract translation: 做出长期稳定信用额度增长决策的方法。 计算机构建包括节点和边缘的马尔可夫决策过程图，其中节点表示一个或多个客户分段的相应状态，并且边缘表示基于历史数据的路径。 计算机对一个或多个客户分段中的相应一个客户分段应用长步骤动作，其中长步骤动作增强多于一个信用限额水平。 计算机计算长步骤动作的获得值。 计算机从长步骤动作中选择最佳的长步骤动作，其中最佳长步骤动作具有最大获得值。

公开(公告)号：US20170140394A1

公开(公告)日：2017-05-18

申请号：US14945389

申请日：2015-11-18

Applicant: International Business Machines Corporation

Inventor： Feng Cao ,  Boliang Chen ,  Andreas Kind ,  Yuan Ni ,  Fei Su ,  Wei Zhao

IPC: G06Q30/02

CPC classification number: G06Q30/0201 ,  H04L9/3236 ,  H04L9/3278 ,  H04L2209/38 ,  H04L2209/56

Abstract: Systems and techniques for blockchain-based reputation tracking in an online marketplace include and/or are configured for: receiving a rate reputation request, the rate reputation request relating to a transaction attempt between at least two entities in the online marketplace; determining a plurality of reputation factors; and determining a reputation adjustment for one or more of the entities participating in the transaction attempt. The reputation adjustment is based at least in part on one or more of the reputation factors, which include: validity of the transaction attempt to which the rate reputation request relates; historical proportion of transactions conducted between the at least two entities participating in the transaction attempt; historical proportion of reputation currency transferred between the entities participating in the transaction attempt; and/or confidence of the validator node with respect to the entities.