磁保持继电器专利分类数据

磁保持继电器专利分类数据，包括磁保持继电器的专利信息，根据人工和算法规则对磁保持继电器的专利信息进行标引、分类，建立多级技术分支分类专题数据库，有助于领域内的创新主体通过工程技术用语快速检索该技术领域内的专利信息。基于分类检索信息，企业可结合专利具体信息进一步分析磁保持继电器专利技术的发展趋势，例如按照年份申请量分析领域发展情况等，引导企业自身研发方向，避免重复研发及专利纠纷产生。1、数据来源：专利检索人员在企业自建数据库中根据关键词、申请人、分类号等条件构建检索式进行检索，得到专利数据。
2、数据清洗：针对检索得到专利数据，对专利的著录项目、标题、说摘要、权利要求书、说明书、说明书附图内容进行摘取与清洗。
3、数据标引：
3.1 技术分支拆解：专利标引人员对专利数据进行技术分支分解，分为多级技术分支。
3.2 技术分支扩展：专利标引人员针对各级技术分支关键词进行概念含义扩展，扩展方向包括：中英文上位、下位、同义词、近义词概念等。
3.3技术分支标引：专利标引人员对专利信息，通过Word2Vec算法（CBOW类型）将专利内容处理为专利词向量数据；在专利内容词向量数据的基础上，利用多类别支持向量机(Multi-class SVM)进行训练，训练建立各级技术分支与各级技术分支关键词以及扩展关键词相对应的专利词向量数据之间映射关系，并基于已训练的多类别支持向量机(Multi-class SVM)模型对单件专利进行技术分支标引，标引其技术分支分类。

Patent classification data for magnetic latching relays includes patent information related to magnetic latching relays. The patent information is indexed and classified according to manual and algorithmic rules, and a multi-level technical branch classification thematic database is established, which helps innovation subjects in the field quickly retrieve patent information in this technical field through engineering technical terms. Based on classified retrieval information, enterprises can further analyze the development trends of magnetic latching relay patent technologies by combining specific patent information, such as analyzing the development status of the field according to annual application volume, etc., to guide the enterprise's own R&D direction and avoid repeated R&D and patent disputes.
1. Data Source: Patent search personnel construct retrieval formulas based on conditions such as keywords, applicants, and classification numbers in the enterprise's self-built database to obtain patent data through retrieval.
2. Data Cleaning: For the retrieved patent data, extract and clean the bibliographic items, title, abstract, claims, specification, and specification drawings of the patents.
3. Data Indexing:
3.1 Technical Branch Decomposition: Patent indexing personnel decompose the patent data into multi-level technical branches.
3.2 Technical Branch Expansion: Patent indexing personnel conduct concept expansion for the keywords of each level of technical branches, including broader terms, narrower terms, synonyms, and near-synonyms in both Chinese and English.
3.3 Technical Branch Indexing: Patent indexing personnel process patent content into patent word vector data using the Word2Vec algorithm (CBOW type) for patent information; based on the patent content word vector data, train using the multi-class Support Vector Machine (Multi-class SVM) to establish the mapping relationship between each level of technical branches and the patent word vector data corresponding to the keywords of each level of technical branches and expanded keywords. Then, perform technical branch indexing on individual patents based on the trained multi-class SVM model to mark their technical branch classifications.

搜集汇总

背景与挑战

背景概述

磁保持继电器专利分类数据包含2972条专利信息，每年更新一次，通过人工和算法规则进行标引和分类，建立了多级技术分支分类专题数据库，有助于企业分析专利技术发展趋势和引导研发方向。