WO2019028620A1 - 基于大数据及人工智能的商标申请的方法 - Google Patents
基于大数据及人工智能的商标申请的方法 Download PDFInfo
- Publication number
- WO2019028620A1 WO2019028620A1 PCT/CN2017/096286 CN2017096286W WO2019028620A1 WO 2019028620 A1 WO2019028620 A1 WO 2019028620A1 CN 2017096286 W CN2017096286 W CN 2017096286W WO 2019028620 A1 WO2019028620 A1 WO 2019028620A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- trademark
- big data
- artificial intelligence
- application
- report
- Prior art date
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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
Definitions
- the invention relates to a method for trademark application, in particular to a method for trademark application based on big data and artificial intelligence.
- an object of the present invention is to provide a method for applying for a trademark application based on big data and artificial intelligence by analyzing a trademark applied for by using a big data algorithm to achieve objective results.
- the present invention adopts the following technical means:
- a method for trademark application based on big data and artificial intelligence comprising the following steps:
- S2 Basic information entry of the trademark to be applied for: the basic information of the trademark to be applied for in S1 is correspondingly entered into the system;
- S4 Extract the graphic of the trademark to be applied to determine whether the same previously registered trademark or the trademark being applied for in the same sub-category: S4.1: If yes, judge the previously registered trademark or is applying Whether the trademark is valid, if it is valid, it is judged whether it is similar to the commodity. If it is approximate, it gives a suggestion that the application is unlikely to be successful and generates a report. If it is not approximate, it gives a suggestion that the application is highly likely to be successful and generates a report.
- the judgment of the non-approximation is judged by the big data module, and the judgment method of the big data module is: classifying according to the grammar rules, each classification rule is judged according to the historical review data, and the results of all the classification rules are in a predetermined manner.
- the weighting value may be adjusted according to the applicant's risk preference, and if it is invalid, determining whether the predetermined time is exceeded, and if it is exceeded, giving a recommendation that the application is unlikely to be successful and generating a report, if not, then Give suggestions for the possibility of successful application and generate a report, S4.2: If not, judge the pending trademark and the previous one Whether the trademark or the trademark being applied for is similar, the similar judgment is judged by the artificial intelligence module, and the artificial intelligence module determines the method: first inputting the graphic, and then analyzing the basic constituent elements, the elements are trademarked
- the graphic elements are basic elements, and the analysis of each basic element composition plan is judged according to the historical review data, and the results of all the classification rules are weighted and calculated according to a certain manner, and the weighted values can be adjusted according to the applicant's risk preference;
- S5 Extract the text of the trademark to be applied to determine whether the same previously registered trademark or the trademark being applied for in the same sub-category:
- S5.1 If yes, judge the previously registered trademark or is applying Whether the trademark is valid, if it is valid, it is judged whether it is similar to the commodity. If it is similar, it gives a suggestion that the application is unlikely to be successful and generates a report. If it is not approximate, it gives a suggestion that the application is highly likely to be successful and generates a report. The judgment of the non-approximation is judged by the big data module, and the judgment method of the big data module is the same as that in S4.1.
- the same in the S4 is determined based on whether the image coincidence degree exceeds a predetermined value.
- the determination of the image coincidence degree includes judging by the coincidence of any angular rotation.
- the value of the trademark to be applied is judged according to the big data: firstly, the dimension is established, including the infringement trial, the transaction amount of the same or similar category trademark, the authorization time, and the trademark of the category. The total number, the rejection rate, and secondly, the weights associated with the transaction amount are given a certain weight, and then weighted to obtain the basic value.
- the coefficient is formed by a factor of 1 greater than 1, and the number of similar trademarks forms a factor 2 with a coefficient less than 1, and the value of the trademark to be applied is equal to the basic value * factor 1 * factor 2.
- the establishment dimension further includes the number of listed companies in the industry and the profitability of listed companies in the industry.
- the establishment dimension further includes the number and rate of registered companies in the industry, the number and rate of e-commerce.
- the infringement trial is to extract the previously registered trademark of the commodity involved in the infringement dispute or the category, city, country or region of the trademark being applied for.
- weights can be dynamically adjusted, and the weights can be adjusted for emerging industries.
- the emerging industry has three criteria: the first is the number of companies registered in the field, the second is the number of companies financing in the field, and the third is Number of company registrations and financing in this area.
- the basic information of the trademark to be applied in S2 is entered into the system by an input module, the information in S4 and S5 is derived from a database, and the information in S4.1, S5.1 and S5.2 is judged by a large data.
- the module analyzes, the information in S4.2 is analyzed by an artificial intelligence module, and the final artificial intelligence module of the trademark to be applied for in S6 is analyzed.
- the output is output by an output module.
- the present invention has the following beneficial effects:
- the OCR technology is used to judge the characters and graphics of the applied trademark, and the OCR technology is an optical character recognition (Optical Character Recognition), which is an optical input method by scanning or the like.
- Optical Character Recognition is an optical input method by scanning or the like.
- the texts of various bills, newspapers, books, manuscripts and other printed materials are converted into image information, and then the image information is transformed into computer input technology that can be used by using text recognition technology.
- the technology is mature and the results of analysis and judgment are guaranteed.
- the registered trademark or the trademark being applied for is analyzed and judged by the big data module and the artificial intelligence module, and the final trademark application analysis report of the trademark to be applied for is output, so that the final value evaluation report of the trademark to be applied is output, and the result reflects the market economic value. It has a high reference value, and the result is more reference value, and the system automatically collects and compares, saves a lot of manpower, and facilitates subsequent monitoring of similar trademarks.
- FIG. 1 is a general flow chart of a method for trademark application based on big data and artificial intelligence according to the present invention
- a method for trademark application based on big data and artificial intelligence includes the following steps:
- S1 Basic information for preparing the trademark to be applied for: the trademark name and design element series of the trademark to be applied for, wherein the design element series includes the content of the text, the information to be highlighted in the pattern, the color, the style, and the like.
- S2 Basic information entry of the trademark to be applied for: Enter the basic information of the trademark to be applied for in S1 into the system.
- the OCR technology is used to judge the text and graphics of the applied trademark.
- the OCR technology is an optical character recognition (Optical Character Recognition), which is a variety of bills, newspapers, books, manuscripts and other printed matter by optical input such as scanning.
- the text is converted into image information, and then the image recognition technology is used to convert the image information into a computer input technology that can be used.
- the technology is mature, and the result of the analysis and judgment is guaranteed. For example, the previously registered trademark or the pending application can be used.
- the information of the trademark is scanned to obtain useful information.
- S4 extracting a graphic of the trademark to be applied, determining whether the same previously registered trademark or the trademark being applied in the same sub-category, the same based on whether the image coincidence degree exceeds a predetermined value, the image coincidence degree
- the judgment includes judging the coincidence of any angle rotation, S4.1: if yes, judging whether the previously registered trademark or the trademark being applied for is valid, and if it is valid, judging whether it is similar to the commodity, if approximate, Give a suggestion that the possibility of successful application is low and generate a report. If it is not approximate, give a suggestion that the possibility of successful application is high and generate a report.
- the judgment of the non-approximation is judged by the big data module, and the judgment method of the big data module To: classify according to grammar rules, each classification rule is judged according to historical review data, and the results of all classification rules are weighted according to a predetermined manner.
- the weighting value may be adjusted according to the applicant's risk preference, if it is invalid, it is judged whether the predetermined time is exceeded, and if it is exceeded, the proposal that the possibility of successful application is given is generated and a report is generated, if not, the report is given A proposal for a high probability of successful application and a report is generated.
- the predetermined value of the image coincidence degree is set to 60%, and if the value of the image coincidence degree is 80%, the predetermined value is exceeded, which gives a low probability of successful application.
- the suggestion and report generation if the value of the image coincidence degree is 40%, then the predetermined value is not exceeded, so that the proposal with high probability of successful application is given and a report is generated.
- the determination of the image coincidence includes judging by the coincidence of any angular rotation. For example, one of the previously registered trademarks or the trademark being applied is vertically placed, and the pattern in the trademark to be evaluated is inclined by 30 degrees. If the angle is placed, then the judgment of the image coincides with the rotation angle. If the image is coincident after 30 degrees of rotation, if the value of the image coincidence value exceeds 60% of the predetermined value, the recommendation that the application success probability is low will be given. Generated newspaper In other words, if the predetermined value is not exceeded, the proposal that the possibility of successful application is given and the report is generated, thus breaking through the difficulty of the graphic to be evaluated, there will be no omission, and the risk coefficient is low.
- Judging the value of the trademark to be applied according to the big data first establish the dimension, including the infringement trial, the transaction amount of the same or similar category trademark, the authorization time, the total number of trademarks of the category, the rejection rate, and the establishment dimension further includes the listed company of the industry.
- the number and the profit rate of listed companies in the industry for example, the number of listed companies in the industry is 1,000, the average profit rate of listed companies in the industry is 20% to 25%, then it can be seen that the profit rate of this industry is still relatively objective.
- Each variable is given a certain weight, and then weighted to obtain the basic value.
- the weighting is formed by a factor of 1 with a coefficient greater than 1, and the number of similar trademarks is formed by a factor of 2 less than 1.
- the value of the trademark to be applied is equal to the basic value.
- the weight of the company type related to the transaction amount is 10%
- the weight of the foreign trade of the commodity is 20%
- the weight of the company's profit margin is 40%
- the weight of the company's total business income is 30%.
- the basic value of the weighted value is 22.14.
- the variables irrelevant to the transaction amount include the company personnel structure, the company established area, the company's survival life, etc.
- the weighting value forming coefficient formed by these variables has a factor of 1.8, so the factor 1 If the factor 2 of the number formation coefficient of a similar trademark is 0.6, the factor 2 is less than 1.
- Such a measurement method assigns weights to different levels to discriminate the differences, and the weighted use does not simply look at one data. Is a comprehensive consideration, the resulting knot The argument is more rational and objective, and the risk factor is small.
- the weights can be adjusted dynamically, and the weights can be adjusted for emerging industries.
- the emerging industry has three criteria: the first is the number of companies registered in the field, the second is the number of companies financing in the field, and the third is the field.
- the number of companies registered and the number of financing this is a common development model in the future business. In this regard, it is more responsive to the trend of the times and is conducive to the subsequent development of the company.
- the infringement trial is the prior registered trademark of the commodity involved in the infringement dispute or the type, city, country or region of the trademark being applied for, the dispute between the previously registered trademark of the infringement dispute or the trademark being applied for.
- the amount and the parties to the dispute so as to avoid disputes in the subsequent development, affect the use of trademarks, and further damage the enterprise.
- S5 Extract the text of the trademark to be applied to determine whether the same prior registered trademark or the trademark being applied for in the same subcategory, the same judgment is judged according to the trademark law, the implementation rules and the examination guideline: S5 .1: If yes, determine whether the previously registered trademark or the trademark being applied for is valid. If it is valid, judge whether it is similar to the commodity. If it is similar, give a suggestion that the application is unlikely to be successful and generate a report. If it is not approximate, it gives a suggestion that the application is highly likely to be successful and generates a report. The judgment of the non-approximation is judged by the big data module, and the judgment method of the big data module is the same as that in S4.1.
- the words and graphics of the applied trademark are respectively judged by OCR technology, and the OCR technology is an abbreviation of optical character recognition (Optical Character) Recognition) converts texts of various bills, newspapers, books, manuscripts and other printed materials into image information through optical input such as scanning, and then uses image recognition technology to convert image information into usable computer input technology.
- OCR technology is an abbreviation of optical character recognition (Optical Character) Recognition) converts texts of various bills, newspapers, books, manuscripts and other printed materials into image information through optical input such as scanning, and then uses image recognition technology to convert image information into usable computer input technology.
- the technology is mature. Provide protection for the results of analysis and judgment.
- the registered trademark or the trademark being applied for is analyzed and judged by the big data module and the artificial intelligence module, and the final trademark application analysis report of the trademark to be applied for is output, so that the final value evaluation report of the trademark to be applied is output, and the result reflects the market economic value. It has a high reference value, and the result is more reference value, and the system automatically collects and compares, saves a lot of manpower, and facilitates subsequent monitoring of similar trademarks.
Abstract
一种基于大数据及人工智能的商标申请的方法，包括以下步骤：S1：准备待申请商标的基本信息，S2：待申请商标的基本信息录入，S3：通过OCR技术分别对待申请商标的文字和图形进行判断，S4：提取待申请商标的图形，判断是否在同一小类有相同的在先已注册的商标或正在申请中的商标，S5：提取待申请商标的文字，判断是否在同一小类有相同的在先已注册的商标或正在申请中的商标，S6：输出待申请商标的商标申请分析报告。上述步骤全部完成后，形成最终的商标申请分析报告，以达到对商标的基本价值进行客观、全面了解以及定位。
Description
本发明涉及一种商标申请的方法，尤指一种基于大数据及人工智能的商标申请的方法。
在日益激烈的市场竞争中,商标的价值越来越得到体现，企业拥有价值高的商标，以及企业越早拥有价值高的专利，便意味着对市场的占有，影响着消费者的消费习惯。
由于商标申请的周期较长，不确定性的因素较多，企业期待能提高商标申请的准确率，然而现在传统的商标在申请时，判断还是以人工为主，不仅存在检索不全面的问题，而且有过多的主观因素，即主观性强，导致无法客观和全面的对待申请商标进行评估分析。
因此，有必要设计一种好的基于大数据及人工智能的商标申请的方法及系统，以克服上述问题。
【发明内容】
针对背景技术所面临的问题，本发明的目的在于提供一种通过采用大数据算法对待申请的商标进行分析，以达到结果客观,参考意义大的基于大数据及人工智能的商标申请的方法。
为实现上述目的，本发明采用以下技术手段：
一种基于大数据及人工智能的商标申请的方法，其包括以下步骤：
S1：准备待申请商标的基本信息：待申请商标的商标名称和设计要素系列资料；
S2：待申请商标的基本信息录入：将S1中待申请商标的基本信息对应录入系统；
S3：通过OCR技术分别对待申请商标的文字和图形进行判断；
S4：提取待申请商标的图形，判断是否在同一小类有相同的在先已注册的商标或正在申请中的商标：S4.1：如果有,则判断在先已注册的商标或正在申请中的商标是否有效，如果是有效,则判断与商品是否近似，如果近似，则给出申请成功可能性低的建议并生成报告，如果不近似,则给出申请成功可能性高的建议并生成报告，所述不近似的判断由大数据模块进行判断，大数据模块的判断方法为：根据语法规则进行分类，每个分类规则,按照历史审查数据进行判断，对所有分类规则的结果按照预定的方式进行加权计算，所述加权值可以根据申请人的风险喜好调整，如果是无效，则判断是否超过预定时间，如果超过,则给出申请成功可能性低的建议并生成报告，如果不超过,则给出申请成功可能性高的建议并生成报告，S4.2：如果无,则判断待申请商标与在先已注册的商标或正在申请中的商标是否相近似，所述相近似的判断由人工智能模块进行判断，人工智能模块的判断方法为：首先将图形输入，接着分析基本组成要素，所述要素以商标的图形要素为基本要素，分析各个基本要素组成方案按照历史审查数据进行判断，对所有分类规则的结果按照一定的方式进行加权计算，所述加权值可以根据申请人的风险喜好调整；
S5：提取待申请商标的文字，判断是否在同一小类有相同的在先已注册的商标或正在申请中的商标：S5.1：如果有，则判断在先已注册的商标或正在申请中的商标是否有效，如果有效，则判断与商品是否近似，如果近似，则给出申请成功可能性低的建议并生成报告，如果不近似，则给出申请成功可能性高的建议并生成报告，所述不近似的判断由大数据模块进行判断，大数据模块的判断方法与S4.1中的相同，如果是无效，则判断是否超过预定时间，如果超过,则给出申请成功可能性低的建议并生成报告，如果不超过,则给出申请成功可能性高的建议并生成报告，S5.2：如果无,则判断待申请商标与在先已注册的商标或正在申请中的商标是否相近似，所述相近似的判断由大数据模块进行判断，大数据模块的判断方法与S4.1中的相同；
S6：输出待申请商标的商标申请分析报告：上述步骤全部完成后，形成最终的商标申请分析报告。
进一步地，S4中的所述相同基于影像重合度是否超过预定值进行判断。
进一步地，所述影像重合度的判断包括任意角度旋转的重合进行判断。
进一步地，S5中的所述相同的判断按照商标法、实施细则及审查指南规定进行判断。
进一步地，S4.1、S5.1和S5.2中，根据大数据判断待申请商标的价值：首先建立维度，包括侵权审判、相同或相近似类别商标的交易金额、授权时间、该类别商标的总体数量、驳回率，其次，对于跟交易金额相关的各变量赋予一定权重，再加权得出基本值。
进一步地，对于跟交易金额无关的各变量加权形成系数大于1的因子1，相似商标的数量形成系数小于1的因子2，待申请商标价值等于基本值*因子1*因子2。
进一步地，建立维度进一步包括该行业上市公司的数量和该行业上市公司的利润率。
进一步地，建立维度进一步包括该行业注册公司的数量及速率、电商的数量及速率。
进一步地，侵权审判为提取该商品涉及侵权纠纷的在先已注册的商标或正在申请中的商标的类别、城市、国家或地区。
进一步地，涉及侵权纠纷的在先已注册的商标或正在申请中的商标的纠纷金钱数额和纠纷当事人。
进一步地，权重可动态调整，对新兴行业可以进行权重调整，新兴行业判断标准为三种：第一种为该领域的公司注册数量，第二种为该领域的公司融资数量，第三种为该领域的公司注册数量和融资数量。
进一步地，S2中的待申请商标的基本信息由一输入模块进行录入系统，S4和S5中的信息来源于一数据库，S4.1、S5.1和S5.2中的信息由一大数据判断模块进行分析，S4.2中的信息由一人工智能模块进行分析，S6中待申请商标的最终一人工智能模块进行分析报
告由一输出模块输出。
与现有技术相比，本发明具有以下有益效果：
上述基于大数据及人工智能的商标申请的方法中，通过OCR技术分别对待申请商标的文字和图形进行判断，OCR技术是光学字符识别的缩写(Optical Character Recognition)，是通过扫描等光学输入方式将各种票据、报刊、书籍、文稿及其它印刷品的文字转化为图像信息，再利用文字识别技术将图像信息转化为可以使用的计算机输入技术，技术成熟，对分析判断的结果提供保障。
首先提取待申请商标的图形，判断是否在同一小类有相同的在先已注册的商标或正在申请中的商标，其次提取待申请商标的文字，判断是否在同一小类有相同的在先已注册的商标或正在申请中的商标，通过大数据模块和人工智能模块进行分析判断，输出待申请商标的最终商标申请分析报告，这样输出待申请商标最终的价值评估报告，结果反映了市场经济价值，具有较高的参考意义，结果也就更具参考价值，并且系统自动采集对比,节省了大量人力，也便于后续类似商标的监控。
图1为本发明基于大数据及人工智能的商标申请的方法的总体流程图；
为便于更好的理解本发明的目的、结构、特征以及功效等，现结合附图和具体实施方式对本发明作进一步说明。
请参见图1，一种基于大数据及人工智能的商标申请的方法，其包括以下步骤：
S1：准备待申请商标的基本信息：待申请商标的商标名称和设计要素系列资料，其中设计要素系列资料包括文字的内容、图案中需要凸显的信息、颜色、风格等。
S2：待申请商标的基本信息录入：将S1中待申请商标的基本信息对应录入系统。
S3：通过OCR技术分别对待申请商标的文字和图形进行判断，OCR技术是光学字符识别的缩写(Optical Character Recognition)，是通过扫描等光学输入方式将各种票据、报刊、书籍、文稿及其它印刷品的文字转化为图像信息，再利用文字识别技术将图像信息转化为可以使用的计算机输入技术，技术成熟，对分析判断的结果提供保障，例如，可以将在先已注册的商标或正在申请中的商标的信息进行扫描，获取有用的信息。
S4：提取待申请商标的图形，判断是否在同一小类有相同的在先已注册的商标或正在申请中的商标，所述相同基于影像重合度是否超过预定值进行判断，所述影像重合度的判断包括任意角度旋转的重合进行判断，S4.1：如果有,则判断在先已注册的商标或正在申请中的商标是否有效，如果是有效,则判断与商品是否近似，如果近似，则给出申请成功可能性低的建议并生成报告，如果不近似,则给出申请成功可能性高的建议并生成报告，所述不近似的判断由大数据模块进行判断，大数据模块的判断方法为：根据语法规则进行分类，每个分类规则,按照历史审查数据进行判断，对所有分类规则的结果按照预定的方式进行加权
计算，所述加权值可以根据申请人的风险喜好调整，如果是无效，则判断是否超过预定时间，如果超过,则给出申请成功可能性低的建议并生成报告，如果不超过,则给出申请成功可能性高的建议并生成报告，例如，影像重合度的预定值设为60％，如果影像重合度的数值为80％，那么就超过了预定值，这样则给出申请成功可能性低的建议并生成报告，如果影像重合度的数值为40％，那么就没有超过了预定值，这样则给出申请成功可能性高的建议并生成报告。所述影像重合的判断包括任意角度旋转的重合进行判断，例如，在先已注册的商标或正在申请中的商标中的一个图案是竖直摆放，而待评估商标中的图案是倾斜30度角的摆放，那么所述影像重合的判断会旋转角度，旋转30度后如果重合，影像重合度的数值只要超过预定值的60％，那么就还是会给出申请成功可能性低的建议并生成报
告，反之，没有超过了预定值，则给出申请成功可能性高的建议并生成报告，这样突破了图形的待评估商标的难点，不会发生遗漏，风险系数低。
根据大数据判断待申请商标的价值：首先建立维度，包括侵权审判、相同或相近似类别商标的交易金额、授权时间、该类别商标的总体数量、驳回率，建立维度进一步包括该行业上市公司的数量和该行业上市公司的利润率，例如，该行业上市公司的数量为1000家，该行业上市公司平均的利润率为20％至25％，那么可以看出这个行业的利润率还是比较客观的，该行业注册公司的数量及速率、电商的数量及速率，如此顺应时代的潮流，这样最终商标的价值才能发挥到最大，后续对获取驰名商标也奠定了基础，其次，对于跟交易金额相关的各变量赋予一定权重，再加权得出基本值，对于跟交易金额无关的各变量加权形成系数大于1的因子1，相似商标的数量形成系数小于1的因子2，待申请商标价值等于基本值*因子1*因子2，S4.2：如果无,则判断待申请商标与在先已注册的商标或正在申请中的商标是否相近似，所述相近似的判断由人工智能模块进行判断，人工智能模块的判断方法为：首先将图形输入，接着分析基本组成要素，所述要素以商标的图形要素为基本要素，分析各个基本要素组成方案按照历史审查数据进行判断，对所有分类规则的结果按照一定的方式进行加权计算，所述加权值可以根据申请人的风险喜好调整。例如，交易金额相关的公司类型的权重为10％，商品对外贸易的权重为20％，公司利润率的权重为40％，公司总营业务收入的权重为30％，从数据库的信息获取，权为100个10，60个20，40个40,80个30，根据加权的计算方法，(10*100+20*60+40*40+30*80)/(100+60+40+80)＝22.14,得出加权的基本值为22.14，交易金额无关的变量有公司人员架构、公司设立的区域、公司的存活寿命等，这些变量形成的加权值形成系数的因子1为1.8,故因子1大于1，相似商标的数量形成系数的因子2为0.6，故因子2小于1，这样的测定方法，对不同层面赋予权重，以判别各差异，配合加权的使用，不会单纯看一个数据，而是综合考虑，这样得出来的结
论是比较理智和客观的，风险系数小。
权重可动态调整，对新兴行业可以进行权重调整，新兴行业判断标准为三种：第一种为该领域的公司注册数量，第二种为该领域的公司融资数量，第三种为该领域的公司注册数量和融资数量，这是以后商业中常见的发展模式，朝这方面考虑，更能顺应时代的潮流，利于企业后续的发展。侵权审判为提取该商品涉及侵权纠纷的在先已注册的商标或正在申请中的商标的类别、城市、国家或地区，涉及侵权纠纷的在先已注册的商标或正在申请中的商标的纠纷金钱数额和纠纷当事人，如此避免后续发展中出现纠纷，影响商标的使用，进一步让企业受损。
S5：提取待申请商标的文字，判断是否在同一小类有相同的在先已注册的商标或正在申请中的商标，所述相同的判断按照商标法、实施细则及审查指南规定进行判断：S5.1：如果有，则判断在先已注册的商标或正在申请中的商标是否有效，如果有效，则判断与商品是否近似，如果近似，则给出申请成功可能性低的建议并生成报告，如果不近似，则给出申请成功可能性高的建议并生成报告，所述不近似的判断由大数据模块进行判断，大数据模块的判断方法与S4.1中的相同，如果是无效，则判断是否超过预定时间，如果超过,则给出申请成功可能性低的建议并生成报告，如果不超过,则给出申请成功可能性高的建议并生成报告，S5.2：如果无,则判断待申请商标与在先已注册的商标或正在申请中的商标是否相近似，所述相近似的判断由大数据模块进行判断，大数据模块的判断方法与S4.1中的相同。
S6：输出待申请商标的商标申请分析报告：上述步骤全部完成后，形成最终的商标申请分析报告。
请参见图1，上述基于大数据及人工智能的商标申请的方法中，通过OCR技术分别对待申请商标的文字和图形进行判断，OCR技术是光学字符识别的缩写(Optical Character
Recognition)，是通过扫描等光学输入方式将各种票据、报刊、书籍、文稿及其它印刷品的文字转化为图像信息，再利用文字识别技术将图像信息转化为可以使用的计算机输入技术，技术成熟，对分析判断的结果提供保障。
首先提取待申请商标的图形，判断是否在同一小类有相同的在先已注册的商标或正在申请中的商标，其次提取待申请商标的文字，判断是否在同一小类有相同的在先已注册的商标或正在申请中的商标，通过大数据模块和人工智能模块进行分析判断，输出待申请商标的最终商标申请分析报告，这样输出待申请商标最终的价值评估报告，结果反映了市场经济价值，具有较高的参考意义，结果也就更具参考价值，并且系统自动采集对比,节省了大量人力，也便于后续类似商标的监控。
以上详细说明仅为本发明之较佳实施例的说明，非因此局限本发明的专利范围，所以，凡运用本创作说明书及图示内容所为的等效技术变化，均包含于本发明的专利范围内。
Claims (10)
- 一种基于大数据及人工智能的商标申请的方法，其特征在于，包括以下步骤：S1：准备待申请商标的基本信息：待申请商标的商标名称和设计要素系列资料；S2：待申请商标的基本信息录入：将S1中待申请商标的基本信息对应录入系统；S3：通过OCR技术分别对待申请商标的文字和图形进行判断；S4：提取待申请商标的图形，判断是否在同一小类有相同的在先已注册的商标或正在申请中的商标：S4.1：如果有,则判断在先已注册的商标或正在申请中的商标是否有效，如果是有效,则判断与商品是否近似，如果近似，则给出申请成功可能性低的建议并生成报告，如果不近似,则给出申请成功可能性高的建议并生成报告，所述不近似的判断由大数据模块进行判断，大数据模块的判断方法为：根据语法规则进行分类，每个分类规则,按照历史审查数据进行判断，对所有分类规则的结果按照预定的方式进行加权计算，所述加权值可以根据申请人的风险喜好调整，如果是无效，则判断是否超过预定时间，如果超过,则给出申请成功可能性低的建议并生成报告，如果不超过,则给出申请成功可能性高的建议并生成报告，S4.2：如果无,则判断待申请商标与在先已注册的商标或正在申请中的商标是否相近似，所述相近似的判断由人工智能模块进行判断，人工智能模块的判断方法为：首先将图形输入，接着分析基本组成要素，所述要素以商标的图形要素为基本要素，分析各个基本要素组成方案按照历史审查数据进行判断，对所有分类规则的结果按照一定的方式进行加权计算，所述加权值可以根据申请人的风险喜好调整；S5：提取待申请商标的文字，判断是否在同一小类有相同的在先已注册的商标或正在申请中的商标：S5.1：如果有，则判断在先已注册的商标或正在申请中的商标是否有效，如果有效，则判断与商品是否近似，如果近似，则给出申请成功可能性低的建议并生成报告，如果不近似，则给出申请成功可能性高的建议并生成报告，所述不近似的判断由大数据模块进行判断，大数据模块的判断方法与S4.1中的相同，如果是无效，则判断是否超过预定时间，如果超过,则给出申请成功可能性低的建议并生成报告，如果不超过,则给出申请成功可能性高的建议并生成报告，S5.2：如果无,则判断待申请商标与在先已注册的商标或正在申请中的商标是否相近似，所述相近似的判断由大数据模块进行判断，大数据模块的判断方法与S4.1中的相同；S6：输出待申请商标的商标申请分析报告：上述步骤全部完成后，形成最终的商标申请分析报告。
- 如权利要求1所述的基于大数据及人工智能的商标申请的方法，其特征在于：S4中的所述相同基于影像重合度是否超过预定值进行判断。
- 如权利要求2所述的基于大数据及人工智能的商标申请的方法，其特征在于：所述影像重合度的判断包括任意角度旋转的重合进行判断。
- 如权利要求1所述的基于大数据及人工智能的商标申请的方法，其特征在于：S5中的所述相同的判断按照商标法、实施细则及审查指南规定进行判断。
- 如权利要求1所述的基于大数据及人工智能的商标申请的方法，其特征在于：S4.1、S5.1和S5.2中，根据大数据判断待申请商标的价值：首先建立维度，包括侵权审判、相同或相近似类别商标的交易金额、授权时间、该类别商标的总体数量、驳回率，其次，对于跟交易金额相关的各变量赋予一定权重，再加权得出基本值。
- 如权利要求5所述的基于大数据及人工智能的商标申请的方法，其特征在于：对于跟交 易金额无关的各变量加权形成系数大于1的因子1，相似商标的数量形成系数小于1的因子2，待申请商标价值等于基本值*因子1*因子2。
- 如权利要求5所述的基于大数据及人工智能的商标申请的方法，其特征在于：建立维度进一步包括该行业上市公司的数量和该行业上市公司的利润率。
- 如权利要求5所述的基于大数据及人工智能的商标申请的方法，其特征在于：建立维度进一步包括该行业注册公司的数量及速率、电商的数量及速率。
- 如权利要求5所述的基于大数据及人工智能的商标申请的方法，其特征在于：侵权审判为提取该商品涉及侵权纠纷的在先已注册的商标或正在申请中的商标的类别、城市、国家或地区；涉及侵权纠纷的在先已注册的商标或正在申请中的商标的纠纷金钱数额和纠纷当事人。
- 如权利要求5所述的基于大数据及人工智能的商标申请的方法，其特征在于：权重可动态调整，对新兴行业可以进行权重调整，新兴行业判断标准为三种：第一种为该领域的公司注册数量，第二种为该领域的公司融资数量，第三种为该领域的公司注册数量和融资数量；S2中的待申请商标的基本信息由一输入模块进行录入系统，S4和S5中的信息来源于一数据库，S4.1、S5.1和S5.2中的信息由一大数据判断模块进行分析，S4.2中的信息由一人工智能模块进行分析，S6中待申请商标的最终一人工智能模块进行分析报告由一输出模块输出。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2017/096286 WO2019028620A1 (zh) | 2017-08-07 | 2017-08-07 | 基于大数据及人工智能的商标申请的方法 |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2017/096286 WO2019028620A1 (zh) | 2017-08-07 | 2017-08-07 | 基于大数据及人工智能的商标申请的方法 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2019028620A1 true WO2019028620A1 (zh) | 2019-02-14 |
Family
ID=65273037
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2017/096286 WO2019028620A1 (zh) | 2017-08-07 | 2017-08-07 | 基于大数据及人工智能的商标申请的方法 |
Country Status (1)
| Country | Link |
|---|---|
| WO (1) | WO2019028620A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110175606A (zh) * | 2019-05-30 | 2019-08-27 | 成都天衡智造科技有限公司 | 一种报告审核分析处理方法及系统 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103500426A (zh) * | 2013-11-07 | 2014-01-08 | 黑龙江慧田知识产权服务有限公司 | 一种适于现代科技的知识产权服务平台 |
| US20140201193A1 (en) * | 2013-01-16 | 2014-07-17 | Wisdomain Inc. | Intellectual property asset information retrieval system |
| CN105426530A (zh) * | 2015-12-15 | 2016-03-23 | 徐庆 | 一种商标检索方法、装置和系统 |
| CN106776541A (zh) * | 2016-11-22 | 2017-05-31 | 北京恒冠网络数据处理有限公司 | 基于大数据撰写专利名称的方法及装置 |
-
2017
- 2017-08-07 WO PCT/CN2017/096286 patent/WO2019028620A1/zh active Application Filing
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140201193A1 (en) * | 2013-01-16 | 2014-07-17 | Wisdomain Inc. | Intellectual property asset information retrieval system |
| CN103500426A (zh) * | 2013-11-07 | 2014-01-08 | 黑龙江慧田知识产权服务有限公司 | 一种适于现代科技的知识产权服务平台 |
| CN105426530A (zh) * | 2015-12-15 | 2016-03-23 | 徐庆 | 一种商标检索方法、装置和系统 |
| CN106776541A (zh) * | 2016-11-22 | 2017-05-31 | 北京恒冠网络数据处理有限公司 | 基于大数据撰写专利名称的方法及装置 |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110175606A (zh) * | 2019-05-30 | 2019-08-27 | 成都天衡智造科技有限公司 | 一种报告审核分析处理方法及系统 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109597994B (zh) | 短文本问题语义匹配方法和系统 | |
| CN107563720A (zh) | 基于大数据及人工智能的商标申请的方法 | |
| CN108090830B (zh) | 一种基于面部画像的信贷风险评级方法和装置 | |
| CN107609022A (zh) | 基于大数据及人工智能的商标申请的系统 | |
| CN111598682A (zh) | 一种针对企业的信用风险评估方法及系统 | |
| TWM582163U (zh) | Face recognition financial planning system | |
| CN107766560B (zh) | 客服服务流程的评价方法和系统 | |
| CN112347254A (zh) | 新闻文本的分类方法、装置、计算机设备和存储介质 | |
| Patil et al. | Offline signature recognition system using histogram of oriented gradients | |
| CN113988459A (zh) | 一种基于电力营销数据的中小企业成长性评估方法及系统 | |
| WO2019028620A1 (zh) | 基于大数据及人工智能的商标申请的方法 | |
| Octiva et al. | Application of The Speed-Up Robust Features Method To Identify Signature Image Patterns On Single Board Computer | |
| US20230206676A1 (en) | Systems and Methods for Generating Document Numerical Representations | |
| Shinde et al. | Feedforward back propagation neural network (FFBPNN) based approach for the identification of handwritten math equations | |
| CN107886233B (zh) | 客服的服务质量评价方法和系统 | |
| CN116343300A (zh) | 人脸特征标注方法、装置、终端以及介质 | |
| WO2019028616A1 (zh) | 基于大数据及人工智能的商标申请的系统 | |
| CN114897322A (zh) | 一种数据资产价值评估系统及方法 | |
| CN114611515A (zh) | 一种基于企业舆情信息识别企业实际控制人的方法和系统 | |
| Blessy et al. | Deep Learning Approach to Offline Signature Forgery Prevention | |
| WO2019028618A1 (zh) | 基于大数据的商标价值评估的方法及系统 | |
| Liang et al. | A Study of Identification of Corporate Financial Fraud Using Neural Network Algorithms in an Information-based Environment | |
| Sadhasivam et al. | Forex exchange using big data analytics | |
| CN111062338A (zh) | 一种证照人像一致性比对方法及其系统 | |
| Meng et al. | The Application Study of Consumer Credit risk model in Auto Financial Institution Based on Logistic Regression |
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: 17920652 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: 17920652 Country of ref document: EP Kind code of ref document: A1 |