面向装备试验鉴定领域的研究热点识别与发展预测

Hotspot Identification and Development Forecasting in the Field of Test Evaluation of Armament

装备试验鉴定是检验武器装备作战效能和作战适用性的重要手段,对其进行研究热点的识别和预测有助于该领域的宏观发展调控。本文针对我国装备试验鉴定领域的发展情况和数据特点,基于论文摘要建立机器学习研究方案。首先通过对比K-Means算法和LDA主题聚类遴选领域研究热点,然后用ARIMA模型进行时间序列预测。结果表明,"武器装备体系作战试验"逐渐成为该领域的研究热点且趋势正盛,与导弹相关的研究仍占据重要地位。本文克服了传统方法对专家知识经验的过分依赖性,为该领域的技术研究方向把控提供有益支撑,也为热点识别的应用研究提供新的思路。

Testing the operational effectiveness and applicability of weapons and equipment contributes to the development and regulation of macro development, so it is very important to identify and predict hot spots in this field. According to the development situation and data characteristics of the equipment test identification field in China, this paper, based on the abstract of the paper establishes the machine learning research solutions,which clusters the subjects of technologies by k-means and LDA. Then, the better result is used for time series prediction through the ARIMA model. The result shows that 'weapon equipment combat test by system of systems' has gradually become the key development object in this field,and missile-related research still occupies an important position.This paper prevents experts from depending on traditional methods, providing beneficial support for the control of technical research direction in this field, and new ideas for the identification of hot research topics.