WO2021128866A1 - 一种专利交易预测方法及系统、专利交易平台 - Google Patents

一种专利交易预测方法及系统、专利交易平台 Download PDF

Info

Publication number
WO2021128866A1
WO2021128866A1 PCT/CN2020/108472 CN2020108472W WO2021128866A1 WO 2021128866 A1 WO2021128866 A1 WO 2021128866A1 CN 2020108472 W CN2020108472 W CN 2020108472W WO 2021128866 A1 WO2021128866 A1 WO 2021128866A1
Authority
WO
WIPO (PCT)
Prior art keywords
transaction
initial
prediction
index
data
Prior art date
Application number
PCT/CN2020/108472
Other languages
English (en)
French (fr)
Inventor
李广凯
郑金
曾森
段力勇
柳勇军
龚余婧
甄春杰
Original Assignee
南方电网科学研究院有限责任公司
保定市大为计算机软件开发有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 南方电网科学研究院有限责任公司, 保定市大为计算机软件开发有限公司 filed Critical 南方电网科学研究院有限责任公司
Priority to US17/789,688 priority Critical patent/US20230052475A1/en
Publication of WO2021128866A1 publication Critical patent/WO2021128866A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
    • G06Q40/04—Trading; Exchange, e.g. stocks, commodities, derivatives or currency exchange
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/18—Legal services
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/18—Legal services
    • G06Q50/184—Intellectual property management

Definitions

  • the invention relates to the field of communication technology, in particular to a patent transaction prediction method and system, and a patent transaction platform.
  • the purpose of the present invention is to provide a patent transaction prediction method and system, and a patent transaction platform to solve the technical problem of the lack of patent transaction trend prediction on the patent transaction platform.
  • the present invention provides a patent transaction prediction method, including the following steps: obtaining patent data to be predicted; constructing a prediction model, which is executed by a computer to predict the transaction probability of the patent data to be predicted; displaying the transaction probability in Among the data attributes of the patent data to be predicted.
  • constructing a prediction model includes: obtaining a data sample, which is a combination of patent data that has occurred transactions; obtaining an initial predictive index of the patent data combination; constructing an initial predictive model of the initial predictive index and transaction probability; To predict the correlation of the forecast model, select the forecast index and weight from the initial forecast index; build a forecast model based on the forecast index and weight.
  • the initial predictive index and the initial predictive model of transaction probability are constructed, including: at least the number of patents in the same family, the number of forward citations, the number of claims, the number of IPCs, the number of inventors, the number of backward citations, the maintenance time, the type of right holder, The straight-line distance between the right holder and the trading platform and the patent transaction price select a number of initial prediction indicators to construct an initial prediction model.
  • selecting the predictive index and the weight from the initial predictive index includes: determining the predictive index based on the initial predictive index and the value of the null hypothesis of the probability of transaction occurrence.
  • the initial prediction model is a logistic regression model.
  • ⁇ 0 in the above formula is a constant term
  • ⁇ 1 to ⁇ i are the coefficients of the independent variables x 1 to x i , respectively.
  • the patent transaction forecasting method of the present invention constructs an initial forecast model from the initial forecast index, and selects the forecast index according to the correlation between the initial forecast index and the initial forecast model, and finally builds the forecast model , Obtain the transaction probability of the patent data to be predicted, realize the prediction of the probability of the patent data being traded, and promote the operation efficiency of the patent transaction of the patent transaction platform.
  • the present invention also provides a patent transaction prediction system, including:
  • the receiving unit is used to obtain the patent data to be predicted; the processing unit is used to construct a prediction model, which is executed by a computer to predict the transaction probability of the patent data to be predicted; and the transaction probability is displayed in the data attribute of the patent data to be predicted .
  • the processing unit is used to: obtain a data sample, which is a combination of patent data of transactions; obtain an initial predictive index of the patent data combination; construct an initial predictive model of the initial predictive index and transaction probability; To predict the correlation of the forecast model, select the forecast index and weight from the initial forecast index; build a forecast model based on the forecast index and weight.
  • the processing unit is used for: at least from the number of patents in the same family, the number of forward citations, the number of claims, the number of IPCs, the number of inventors, the number of backward citations, the maintenance time, the type of the right holder, the straight-line distance between the right holder and the trading platform, and
  • the patent transaction price selects a number of initial predictive indicators to construct an initial predictive model; the predictive index is determined according to the initial predictive index and the value of the null hypothesis of the occurrence of the transaction probability.
  • the present invention also provides a patent transaction platform, including the above-mentioned patent transaction prediction system.
  • Figure 1 is a flowchart of the patent prediction method of the present invention
  • Fig. 2 is a flow chart of constructing a prediction model of the present invention
  • FIG. 3 is a block diagram of the patent prediction system of the present invention.
  • Receiving unit 12. Processing unit.
  • the "plurality” mentioned in this embodiment refers to two or more.
  • “And/or” describes the association relationship of the associated objects, indicating that there can be three types of relationships, for example, A and/or B, which can mean: A alone exists, A and B exist at the same time, and B exists alone.
  • Words such as "exemplary” or “for example” are used as examples, illustrations, or illustrations, and are intended to present related concepts in a specific manner, and should not be construed as being more preferred or advantageous than other embodiments or design solutions.
  • IPC International Patent Classification
  • the present invention provides a patent transaction prediction method and system, and a patent transaction platform.
  • a patent transaction prediction method provided by the present invention includes the following steps:
  • the patent transaction prediction method of the present invention is applied to a patent transaction platform.
  • the patent data information contains the initial predictive index of the patent. Including: transaction price, distance between right holder and trading platform, right maintenance time, right holder type, number of claims, number of inventors, number of forward citations, number of IPCs, number of inventors, number of backward citations, and number of patents of the same family. It should be understood that when constructing different patent prediction models, it may not be limited to the above prediction and prediction indicators.
  • the initial prediction index is used to construct the initial prediction model, and the prediction index is selected through the analysis of the initial prediction index to construct the prediction model, and the patent data to be predicted is selected to realize the prediction of the transaction probability of the patent data to be predicted.
  • Patent transactions include changes in the legal status of patents, including: patent authorization, patent licensing, patent transfer, patent pledge, and patent invalidation. It should be understood that patent transactions are not limited to changes in the legal status of the aforementioned patents.
  • the value of the transaction probability is displayed in the attribute of the patent data to be predicted.
  • the displayed patent transaction probability will increase the probability of the transaction of the patent data to be predicted.
  • the prediction model is used to predict the transaction probability of the patent data to be predicted, and the probability is displayed in the data attribute of the patent data, so as to increase the probability of the patent data transaction to be predicted.
  • constructing a prediction model includes:
  • initial prediction model of initial prediction indicators and transaction probability including: at least from the number of patents in the same family, the number of forward citations, the number of claims, the number of IPCs, the number of inventors, and the number of backward citations. , Maintenance time, right holder type, straight-line distance between right holder and trading platform, and patent transaction price Select a number of initial predictive indicators to construct an initial predictive model.
  • the patent transaction price shows the patent holder’s expectation of the value of the patent; the straight-line distance between the right holder and the trading platform affects the cost of identification and supervision of the trading platform; the patent maintenance time means the patent from the date of application to invalidation, termination, and revocation Or the actual time on the expiration date; the types of right holders include: individuals, enterprises, high-efficiency and scientific research institutions; the number of forward citations is the number of times a patent is cited by a later patent; the IPC number is the number of international patent classification numbers in the patent document; The number of citations refers to the number of previous patent documents cited in patent application documents; the number of patents in the same family refers to the number of patents with common priority filed and published by patentees in different countries or regions.
  • the predictive index and the weight are selected from the initial predictive index, including: according to the initial predictive index and the rejection of the null hypothesis of the transaction probability The value determines the predictor.
  • this initial predictive index is selected as the predictive index.
  • the initial prediction model is a logistic regression model.
  • the logistic regression model is selected as the binary logistic regression model.
  • ⁇ 0 in the above formula is a constant term
  • ⁇ 1 to ⁇ i are the coefficients of the independent variables x 1 to x i , respectively.
  • the present invention also provides a patent transaction prediction system, including: a receiving unit 11 for obtaining patent data to be predicted; a processing unit 12 for constructing a prediction model, which is executed by a computer to predict the transaction probability of the patent data to be predicted ; And display the transaction probability in the data attribute of the patent data to be predicted.
  • the processing unit 12 is used to: obtain data samples, which are patent data combinations of transactions that have occurred; obtain initial predictive indicators of patent data combinations; construct initial predictive indicators and initial transaction probabilities Forecasting model; according to the correlation between the initial forecasting index and the initial forecasting model, selecting the forecasting index and weight from the initial forecasting index; constructing the forecasting model based on the forecasting index and the weighting.
  • the processing unit 12 is used for: at least from the number of patents in the same family, the number of forward citations, the number of claims, the number of IPCs, the number of inventors, the number of backward citations, the maintenance time, the type of right holder, The straight-line distance between the right holder and the trading platform and the patent transaction price select a number of initial predictive indicators to construct an initial predictive model; the predictive index is determined based on the initial predictive index and the value of the null hypothesis of the transaction probability.
  • the present invention also provides a patent transaction platform, including the above-mentioned patent transaction prediction system.
  • the patent transaction prediction method of the present invention also provides a specific embodiment, which is specifically as follows:
  • the present invention chooses to analyze the probability of patent transactions from the perspectives of changes in the state of use of the rights due to the transfer of patent rights and changes in the state of invalidation of the rights due to non-payment of annual fees by the patentee.
  • the selected initial predictive indicators include the patent transaction price of the right holder, the distance between the right holder and the patent transaction platform, the number of patents in the same family, the number of forward citations, the number of claims, the number of IPCs, the number of inventors, the number of backward citations, and the maintenance time.
  • the dependent variable patent legal status changes or not is a binary variable
  • the binary Logistic model can be used for regression analysis.
  • the value of the dependent variable y in the regression model is 1, which means that the legal status of A or B has changed; the value of y is 0, which means that no corresponding legal status changes have occurred.
  • the function P represents the probability of a change in the legal state of A or B.
  • the independent variables in the function P are denoted as x 1 , x 2 ,..., x i , and a Logistic regression model that estimates the probability of a change in the legal state can be obtained.
  • ⁇ 0 is a constant term
  • ⁇ 1 to ⁇ i are the regression coefficients of the independent variables x 1 to x i, respectively.
  • the forward screening strategy is used to gradually introduce independent variables into the regression equation until no more statistically significant independent variables can be introduced.
  • 6 independent variables enter the regression equation: forward reference Number, number of claims, number of IPC, number of backward citations, listed price, distance between the right holder and the exchange.

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Tourism & Hospitality (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Technology Law (AREA)
  • Development Economics (AREA)
  • Operations Research (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Game Theory and Decision Science (AREA)
  • Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • General Health & Medical Sciences (AREA)
  • Quality & Reliability (AREA)
  • Educational Administration (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)

Abstract

一种专利交易预测方法，包括以下步骤：获取待预测专利数据（S1）；构建预测模型，预测模型由计算机执行，以预测待预测专利数据的交易概率（S2）；将交易概率显示在待预测专利数据的数据属性中（S3）。该方法将专利交易概率显示在专利数据属性中，提高专利交易概率。还提供一种专利交易预测系统、专利交易平台，同样具有上述优点。

Description

一种专利交易预测方法及系统、专利交易平台 技术领域
本发明涉及通讯技术领域，具体涉及一种专利交易预测方法及系统、专利交易平台。
背景技术
目前我国技术市场成交量增长迅速，利用互联网开展技术交易等科技服务具有巨大潜力。发展在线化交易服务技术的目的在于降低交易成本、缓解交易过程的信息不对称、提高服务协同共享能力。
在专利交易过程中，平台管理者缺少对专利技术交易趋势、识别技术交易的潜力的预测，导致网上专利技术运作效率低。
发明内容
本发明的目的在于提供一种专利交易预测方法及系统、专利交易平台，以解决专利交易平台上缺乏对专利交易趋势预测这一技术问题。
为达到上述目的，本发明提供一种专利交易预测方法，包括以下步骤：获取待预测专利数据；构建预测模型，预测模型由计算机执行，以预测待预测专利数据的交易概率；将交易概率显示在待预测专利数据的数据属性中。
优选地，构建预测模型包括：获取数据样本，数据样本为已发生交易的专利数据组合；获取专利数据组合的初始预测指标；构建初始预测指标和交易概率的初始预测模型；根据初始预测指标与初始预测模型的相关度，从初始预测指标中选取预测指标，以及权重；基于预测指标和权重构建预测模型。
优选地，构建初始预测指标和交易概率的初始预测模型，包括：至少从同族专利数、前向引用数、权利要求数、IPC数、发明人数、后向引用数、维持时间、权利人类型、权利人与交易平台的直线距离和专利交易价格选取若干个初始预测指标，构建初始预测模型。
优选地，根据初始预测指标与初始预测模型的相关度，从初始预测指标中选取预测指标，以及权重，包括：根据初始预测指标与交易概率发生的拒绝原假设的值确定预测指标。
优选地，初始预测模型为logistic回归模型。
优选地，logistic回归模型中专利数据发生交易的概率为P(y i＝1│x 1,x 2,…,x i)，其中P满足以下公式：
Figure PCTCN2020108472-appb-000001
Figure PCTCN2020108472-appb-000002
其中，上式中β 0为常数项，β 1～β i分别为自变量x 1～x i的系数。
与现有技术相比，本发明的专利交易预测方法，通过从初始预测指标中构建初始预测模型，并根据初始预测指标与初始预测模型之间的相关 度，选取预测指标，并最终构建预测模型，获得待预测专利数据的交易概率，实现对专利数据被交易的概率的预测，推动了专利交易平台专利交易的运作效率。
本发明还提供一种专利交易预测系统，包括：
接收单元，用于获取待预测专利数据；处理单元，用于构建预测模型，预测模型由计算机执行，以预测待预测专利数据的交易概率；并将交易概率显示在待预测专利数据的数据属性中。
优选地，处理单元用于：获取数据样本，数据样本为已发生交易的专利数据组合；获取专利数据组合的初始预测指标；构建初始预测指标和交易概率的初始预测模型；根据初始预测指标与初始预测模型的相关度，从初始预测指标中选取预测指标，以及权重；基于预测指标和权重构建预测模型。
优选地，处理单元用于：至少从同族专利数、前向引用数、权利要求数、IPC数、发明人数、后向引用数、维持时间、权利人类型、权利人与交易平台的直线距离和专利交易价格选取若干个初始预测指标，构建初始预测模型；根据初始预测指标与交易概率发生的拒绝原假设的值确定预测指标。
与现有技术相比，本发明提供的专利交易预测系统的有益效果与上述专利交易预测方法的有益效果相同，在此不做赘述。
本发明还提供一种专利交易平台，包括上述专利交易预测系统。
与现有技术相比，本发明提供的专利交易平台的有益效果与上述专利交易预测方法的有益效果相同，在此不做赘述。
附图说明
图1为本发明的专利预测方法的流程图；
图2为本发明的构建预测模型的流程图；
图3为本发明的专利预测系统的框图；
附图标记：
11.接收单元、12.处理单元。
具体实施方式
下面结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
本实施例中提到的“多个”是指两个或两个以上。“和/或”，描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B的情况。“示例性的”或者“例如”等词用于表示作例子、例证或说明，旨在以具体方式呈现相关概念，不应被解释为比其他实施例或设计方案更优选或更具优势。
在介绍本申请实施例之前首先对本申请实施例中涉及到的相关名词作如下释义：
国际专利分类法(international patent classification，缩写为IPC)。
随着社会的发展，越来越多的专利交易平台或管理系统涌现，专利被当做商品被交易，当也应当注意到，由于专利交易平台上缺少对专利交易预测，致使专利管理平台上大量专利被闲置。
为解决上述技术问题，本发明提供一种专利交易预测方法及系统、专利交易平台。
如图1所述，本发明提供的一种专利交易预测方法，包括以下步骤：
S1.获取待预测专利数据；
需要说明的是，本发明的专利交易预测方法应用在专利交易平台上。专利交易平台上有若干专利数据，获取待预测专利数据。此时专利数据信息中包含有专利的初始预测指标。包括：交易价格、权利人与交易平台的距离、权利维持时间、权利人类型、权利要求数量、发明人数、前向引用数、IPC数、发明人数、后向引用数以及同族专利数。应理解，当构建不同的专利预测模型，可不限于上述预测预测指标。
S2.构建预测模型，预测模型由计算机执行，以预测待预测专利数据的交易概率。
在本实施例中，利用初始初测指标构建初始预测模型，并通过对初始预测指标的分析选取预测指标构建预测模型，选取待预测专利数据后实现对待预测专利数据的交易概率的预测。
专利交易包括专利法律状态发生变化，包括：专利授权、专利许可、专利转让、专利质押、专利失效。应理解，专利交易不限于上述专利法律状态发生变化。
S3.将交易概率显示在待预测专利数据的数据属性中。
将交易概率的数值显示在待预测专利数据的属性中，待用户在专利交易平台浏览时，显示的专利交易概率会提高待预测专利数据交易的可能性。
采用上述技术方案，采用预测模型预测待预测专利数据的交易概率并将概率显示在专利数据的数据属性中，提高待预测专利数据交易的概率。
在上述实施例的基础上，进一步地，构建预测模型包括：
S20.获取数据样本，数据样本为已发生交易的专利数据组合；
S21.获取专利数据组合的初始预测指标；
S22.构建初始预测指标和交易概率的初始预测模型；
S23.根据初始预测指标与初始预测模型的相关度，从初始预测指标中选取预测指标，以及权重；
S24.基于预测指标和权重构建预测模型。
在上述实施例的基础上，进一步地，构建初始预测指标和交易概率的初始预测模型，包括：至少从同族专利数、前向引用数、权利要求数、IPC数、发明人数、后向引用数、维持时间、权利人类型、权利人与交易平台的直线距离和专利交易价格选取若干个初始预测指标，构建初始预测模型。
需要说明的是，专利交易价格显示了专利权人对专利价值的预期；权利人与交易平台的直线距离影响交易平台的识别与监督成本；专利维持时间表示专利从申请日至无效、终止、撤销或届满之日的实际时间；权利人类型包括：个人、企业、高效和科研机构；前向引用数，为专利被后期专利引用的次数；IPC数是专利文件中国际专利分类号的数量；后向引用数，为专利申请文件中引用前人专利文献的数量；同族专利数，是专利权人在不同国家或地区申请、公布的具有共同优先权的专利数量。
在上述实施例的基础上，进一步地，根据初始预测指标与初始预测模型的相关度，从初始预测指标中选取预测指标，以及权重，包括：根据初始预测指标与交易概率发生的拒绝原假设的值确定预测指标。
需要说明的是，若初始预测指标与交易概率的拒绝原假设的值小于0.05，则将此初始预测指标选为预测指标。
在上述实施例的基础上，进一步地，初始预测模型为logistic回归模型。
在本实施例中选择logistic回归模型为二分类logistic回归模型。
在上述实施例的基础上，进一步地，logistic回归模型中专利数据发生交易的概率为P(y i＝1│x 1,x 2,…,x i)，其中P满足以下公式：
Figure PCTCN2020108472-appb-000003
Figure PCTCN2020108472-appb-000004
其中，上式中β 0为常数项，β 1～β i分别为自变量x 1～x i的系数。
本发明还提供一种专利交易预测系统，包括：接收单元11，用于获取待预测专利数据；处理单元12，用于构建预测模型，预测模型由计算机执行，以预测待预测专利数据的交易概率；并将交易概率显示在待预测专利数据的数据属性中。
在上述实施例的基础上，进一步地，处理单元12用于：获取数据样本，数据样本为已发生交易的专利数据组合；获取专利数据组合的初始预测指标；构建初始预测指标和交易概率的初始预测模型；根据初始预测指标与初始预测模型的相关度，从初始预测指标中选取预测指标，以及权重；基于预测指标和权重构建预测模型。
在上述实施例的基础上，进一步地，处理单元12用于：至少从同族专利数、前向引用数、权利要求数、IPC数、发明人数、后向引用数、维持时间、权利人类型、权利人与交易平台的直线距离和专利交易价格选取若干个初始预测指标，构建初始预测模型；根据初始预测指标与交易概率发生的拒绝原假设的值确定预测指标。
本发明还提供一种专利交易平台，包括上述专利交易预测系统。
本发明的专利交易预测方法还提供一具体实施例，具体如下：
选取某网上技术交易平台的挂牌专利，确定IPC分类号A61大类的挂牌专利作为分析对象，共计获得A61大类下的样本87件，其中，维持专利有效状态的样本数为15件，权利转让状态发生变化的专利为42件，专利失效状态的专利为30件。
本发明选择从专利权转让产生权利运用状态的变化、专利权人未缴年费产生权利失效状态的变化两个角度，分析专利交易概率，即：法律状态变化的影响因素，建立相关预测模型。
为了构建两种专利法律状态变化的预测模型，确定了网上专利交易中法律状态变化的两个主要预测对象，分别是A：专利权转让产生权利运用状态的变化；B：专利权人未缴年费产生权利失效状态的变化，并将相应法律状态变化与否作为统计分析的因变量。
选取的初始预测指标包括权利人的专利交易价格、权利人与专利交易平台的距离、同族专利数、前向引用数、权利要求数、IPC数、发明人数、后向引用数和维持时间。
因变量专利法律状态变化与否是二分类变量，可利用二分类Logistic模型进行回归分析。回归模型中的因变量y取值为1，表示发生A或B的法律状态变化；y取值为0，表示未发生相应的法律状态变化。函数P代表发生A或B法律状态变化的概率，函数P中的自 变量分别记为x 1，x 2，…，x i，则可得到估计发生法律状态变化概率的Logistic回归模型。
专利发生某一专利法律状态变化的概率P(y i＝1|x 1，x 2，...，x i)可以表示为：
Figure PCTCN2020108472-appb-000005
Figure PCTCN2020108472-appb-000006
上式中β 0为常数项，β 1～β i分别为自变量x 1～x i的回归系数。
对于因变量A回归分析，使用向前筛选策略逐步将自变量引入回归方程，直至再无具有统计学意义的自变量可被引入时为止，最终进入回归方程的自变量为6个：前向引用数、权利要求数、IPC数、后向引用数、挂牌价格、权利人与交易所距离。
由回归模型A的结果，预测专利i发生专利转让的概率P为：
Figure PCTCN2020108472-appb-000007
类似地，对于因变量B回归分析，最终进入回归方程的自变量为5个：前向引用数、权利要求数、后向引用数、发明人数、维持时间。
由回归模型B的结果，预测专利i未缴年费而失效的概率P为：
Figure PCTCN2020108472-appb-000008
上面所述的实施例仅仅是对本发明的优选实施方式进行描述，并非对本发明的构思和范围进行限定。在不脱离本发明设计构思的前提下，本领域普通人员对本发明的技术方案做出的各种变型和改进，均应落入到本发明的保护范围，本发明请求保护的技术内容，已经全部记载在权利要求书中。

Claims (10)

  1. 一种专利交易预测方法，应用于专利交易平台，其特征在于，包括以下步骤：
    获取待预测专利数据；
    构建预测模型，所述预测模型由计算机执行，以预测所述待预测专利数据的交易概率；
    将所述交易概率显示在所述待预测专利数据的数据属性中。
  2. 根据权利要求1所述的专利交易预测方法，其特征在于，所述构建预测模型包括：
    获取数据样本，所述数据样本为已发生交易的专利数据组合；
    获取所述专利数据组合的初始预测指标；
    构建所述初始预测指标和所述交易概率的初始预测模型；
    根据所述初始预测指标与所述初始预测模型的相关度，从所述初始预测指标中选取预测指标，以及权重；
    基于所述预测指标和权重构建所述预测模型。
  3. 根据权利要求1或2所述的专利交易预测方法，其特征在于，所述构建所述初始预测指标和所述交易概率的初始预测模型，包括：
    至少从同族专利数、前向引用数、权利要求数、IPC数、发明人数、后向引用数、维持时间、权利人类型、权利人与所述交易平台的直线距离和专利交易价格选取若干个所述初始预测指标，构建所述初始预测模型。
  4. 根据权利要求1所述的专利交易预测方法，其特征在于，所述根据所述初始预测指标与所述初始预测模型的相关度，从所述初始预测指标中选取预测指标，以及权重，包括：
    根据所述初始预测指标与所述交易概率发生的拒绝原假设的值确定所述预测指标。
  5. 根据权利要求4所述的专利交易预测方法，其特征在于，所述初始预测模型为logistic回归模型。
  6. 根据权利要求5所述的专利交易预测方法，其特征在于，所述logistic回归模型中专利数据发生交易概率为P(y i＝1|x 1,x 2,…,x i)，其中P满足以下公式：
    Figure PCTCN2020108472-appb-100001
    Figure PCTCN2020108472-appb-100002
    其中，上式中β 0为常数项，β 1～β i分别为自变量x 1～x i的系数。
  7. 一种专利交易预测系统，其特征在于，包括：
    接收单元，用于获取待预测专利数据；
    处理单元，用于构建预测模型，所述预测模型由计算机执行，以预测所述待预测专利数据的交易概率；并将所述交易概率显示在所述待预测专利数据的数据属性中。
  8. 根据权利要求7所述的专利交易预测系统，其特征在于，所述处理单元用于：
    获取数据样本，所述数据样本为已发生交易的专利数据组合；
    获取所述专利数据组合的初始预测指标；
    构建所述初始预测指标和所述交易概率的初始预测模型；
    根据所述初始预测指标与所述初始预测模型的相关度，从所述初始预测指标中选取预测指标，以及权重；
    基于所述预测指标和权重构建所述预测模型。
  9. 根据权利要求8所述的专利交易预测系统，其特征在于，所述处理单元用于：
    至少从同族专利数、前向引用数、权利要求数、IPC数、发明人数、后向引用数、维持时间、权利人类型、权利人与所述交易平台的直线距离和专利交易价格选取若干个所述初始预测指标，构建所述初始预测模型；
    根据所述初始预测指标与所述交易概率发生的拒绝原假设的值确定所述预测指标。
  10. 一种专利交易平台，其特征在于，包含权利要求6-8任一项所述的专利交易预测系统。
PCT/CN2020/108472 2019-12-28 2020-08-11 一种专利交易预测方法及系统、专利交易平台 WO2021128866A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/789,688 US20230052475A1 (en) 2019-12-28 2020-08-11 Patent transaction prediction method and system, and patent transaction platform

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911384503.2 2019-12-28
CN201911384503.2A CN113052358A (zh) 2019-12-28 2019-12-28 一种专利交易预测方法及系统、专利交易平台

Publications (1)

Publication Number Publication Date
WO2021128866A1 true WO2021128866A1 (zh) 2021-07-01

Family

ID=76507688

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/108472 WO2021128866A1 (zh) 2019-12-28 2020-08-11 一种专利交易预测方法及系统、专利交易平台

Country Status (3)

Country Link
US (1) US20230052475A1 (zh)
CN (1) CN113052358A (zh)
WO (1) WO2021128866A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103813292A (zh) * 2014-01-16 2014-05-21 中山大学 基于Logistic回归法信用预测的频谱交易定价策略
CN107330741A (zh) * 2017-07-07 2017-11-07 北京京东尚科信息技术有限公司 分品类电子券使用预测方法、装置及电子设备
CN110047002A (zh) * 2019-03-28 2019-07-23 莆田学院 一种基于数据分析的期货推荐方法及系统
CN110059896A (zh) * 2019-05-15 2019-07-26 浙江科技学院 一种基于强化学习的股票预测方法及系统
CN110415119A (zh) * 2019-07-30 2019-11-05 中国工商银行股份有限公司 模型训练、票据交易预测方法、装置、存储介质及设备

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103813292A (zh) * 2014-01-16 2014-05-21 中山大学 基于Logistic回归法信用预测的频谱交易定价策略
CN107330741A (zh) * 2017-07-07 2017-11-07 北京京东尚科信息技术有限公司 分品类电子券使用预测方法、装置及电子设备
CN110047002A (zh) * 2019-03-28 2019-07-23 莆田学院 一种基于数据分析的期货推荐方法及系统
CN110059896A (zh) * 2019-05-15 2019-07-26 浙江科技学院 一种基于强化学习的股票预测方法及系统
CN110415119A (zh) * 2019-07-30 2019-11-05 中国工商银行股份有限公司 模型训练、票据交易预测方法、装置、存储介质及设备

Also Published As

Publication number Publication date
US20230052475A1 (en) 2023-02-16
CN113052358A (zh) 2021-06-29

Similar Documents

Publication Publication Date Title
Bourdet et al. Emigrants' remittances and Dutch disease in Cape Verde
Camanho et al. Cost efficiency, production and value-added models in the analysis of bank branch performance
Cheng et al. Alternative approach to credit scoring by DEA: Evaluating borrowers with respect to PFI projects
Hensher et al. Forecasting corporate bankruptcy: Optimizing the performance of the mixed logit model
CN112862298B (zh) 一种针对用户画像的信用评估方法
Sepehrdoust et al. The knowledge-based products and economic complexity in developing countries
Baek et al. A technology valuation model to support technology transfer negotiations
WO2020143345A1 (zh) 仓单质押的信贷风险监控方法及装置
Wu et al. A novel two-stage method for matching the technology suppliers and demanders based on prospect theory and evidence theory under intuitionistic fuzzy environment
Thomas A relationship between technology indicators and stock market performance
Xuan Regression analysis of supply chain financial risk based on machine learning and fuzzy decision model
CN108734567A (zh) 一种基于大数据人工智能风控的资产管理系统及其评估方法
CN110489691A (zh) 页面组件显示方法及终端设备
Nguyen et al. Spherical Fuzzy WASPAS-based Entropy Objective Weighting for International Payment Method Selection.
Tavana et al. Analytic hierarchy process and data envelopment analysis: A match made in heaven
CN114254846A (zh) 一种工程项目信息管理系统
Jiang Application and comparison of multiple machine learning models in finance
WO2021128866A1 (zh) 一种专利交易预测方法及系统、专利交易平台
Saleemi In COVID-19 outbreak, correlating the cost-based market liquidity risk to microblogging sentiment indicators
CN116091242A (zh) 推荐产品组合生成方法及装置、电子设备和存储介质
JP2002015043A (ja) 鮮魚の流通支援情報の提供方法および鮮魚の流通支援情報の提供装置
Song Design and application of financial market option pricing system based on high-performance computing and deep reinforcement learning
Teschner et al. Evaluating hidden market design
Yu Linking the balanced scorecard to business models for value-based strategic management in e-Business
Xu et al. A new approach to decision-making with key constraint and its application in enterprise information systems

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20906595

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20906595

Country of ref document: EP

Kind code of ref document: A1