衢州市链主企业研发需求与高校专家人才智能匹配数据

适用范围：衢州市链主企业
1.主动挖掘企业技术需求：可以根据自身技术研发，主动挖掘企业下一步的技术需求。
2.为企业匹配技术专家：解决链主企业和高校人才之间的匹配问题，帮助企业找对应的技术专家。
3.选择技术方向：匹配高校专家，深入企业调研，协助企业确定技术方向。
4.促成企业产学研合作：为有需要借助外力研发的企业提供渠道，促进链主企业的产学研合作。
5.创建高校科技成果转化。为高校老师链接有需求的企业，促进科技成果转化。1.定期采集更新企业专利数据：发明公布专利、发明授权专利、实用新型专利和外观设计专利（发明公布和发明授权是不同阶段的企业专利状态，不会重复）
2.进行专利数据清洗，并根据内部系统的技术分类体系为专利加标签，采用技术分类关键词匹配专利描述，将每个匹配到的关键词自动添加为企业技术标签
3.计算专利标签集中度，关键词每命中一次i+1，采用命中次数maxΣi 最高的前5个标签为企业标签
4.采集高校专家人才信息，并根据人才信息（姓名+单位）采集专家人才专利数据，采用步骤2和3的方法分析专利数据并为人才添加标签（含关键词命中次数）
5.标签匹配：根据企业标签匹配高校人才标签，为企业推荐命中关键词匹配次数最高（按照企业单个标签次数与人才单个标签之和取数）的前10名高校专家（少于10名则全部展示）

Scope of Application: Leading Chain Enterprises in Quzhou City

1. Proactive mining of enterprise technological needs: Proactively identify the next-stage technological demands of enterprises based on their own R&D activities.
2. Matching technical experts for enterprises: Address the matching gap between leading chain enterprises and university talents, and help enterprises find suitable technical experts.
3. Technical direction selection: Collaborate with matched university experts to conduct in-depth on-site investigations at enterprises and assist in determining their technical directions.
4. Facilitating industry-university-research cooperation: Provide channels for enterprises in need of external R&D support, and promote industry-university-research cooperation among leading chain enterprises.
5. Facilitating university scientific and technological achievement transformation: Connect university faculty with enterprises in need, so as to promote the transformation of scientific and technological achievements.

1. Regular collection and update of enterprise patent data: Including invention publication patents, invention authorization patents, utility model patents, and design patents (invention publication and invention authorization represent different stages of patent status and will not be duplicated).
2. Patent data cleaning: Add tags to patents based on the technical classification system of the internal system, match patent descriptions with technical classification keywords, and automatically add each matched keyword as an enterprise technical tag.
3. Calculation of patent tag concentration: For each hit of a keyword, increment the count by 1; select the top 5 tags with the highest total hit counts (max Σi) as the enterprise's official technical tags.
4. Collection of university expert talent information: Collect patent data of experts based on their personal information (name + affiliated institution), then analyze the patent data using the methods described in steps 2 and 3, and add tags to the experts (including keyword hit counts).
5. Tag matching: Match enterprise tags with university talent tags, and recommend the top 10 university experts with the highest total keyword matching counts (calculated as the sum of hit counts for each individual tag between the enterprise and the expert) to the enterprise; all eligible experts will be recommended if fewer than 10 exist.