基于多层感知机的技术创新人才发现方法

专利数据中包含了丰富的科技信息。通过对专利数据的分析,不仅可以帮助企业快速了解当前领域的发展状况,还可以及时发现领域中的技术创新人才。当前的技术创新人才发现方法往往是孤立地从各个专利特征的角度来评估发明人的技术创新实力,而没有统一的学习模型来发现专利集合中的技术创新人才。针对该问题,提出一种基于多层感知机模型的技术创新人才发现方法。从专利数据中抽取发明人的各个特征;基于抽取的发明人特征构建多层感知机模型;采用该多层感知机模型在专利数据集合中发现技术创新人才。实验结果表明,该方法是有效的。

A TECHNICAL INNOVATION TALENTS DISCOVERY METHOD BASED ON MULTI-LAYER PERCEPTRON

Patent data contains rich scientific and technological information. The analysis of patent data not only can help enterprises quickly understand the current situation of development in the field, but also can timely discover the technological innovation talents in the field. Existing methods for technical innovation talents discovery generally evaluate the strength of the inventors by various patent features in isolation, and there is not a unified learning model to discover the technical innovation talents in the patent collection. In view of this, we proposed a technical innovation talents discovery method based on multi-layer perceptron. We extracted the inventor s features from patent data, constructed a multi-layer perceptron model based on the extracted features, and used the multi-layer perceptron model to discover technical innovation talents in the patent collection. The experimental results show that the proposed method is effective.