[目的/意义]探究中国政府网"一带一路"相关新闻的话题内容及热度变化，呈现"一带一路"倡议主题及动态，明确不同时期的倡议重点，为相关研究提供参考。[方法/过程]构建基于LDA模型的新闻话题内容的基本框架，限定2015-2017年"一带一路"相关新闻数据，利用LDA模型进行话题抽取，根据文档与话题的概率分布计算，分析各主题在不同时间段的热度演化。[结果/结论]抽取得出30个细分话题，归纳为政策沟通、设施联通、贸易畅通、资金融通、民心相通、"一带一路"对我国经济的影响和政府工作7大类。其中，政策沟通类在全时间段上热度最高，贸易畅通和"一带一路"对我国经济的影响两类话题紧随其后。"进出口"等细分话题的热度不断上升，"改革与转型"等细分话题的热度则有下降，体现了官方媒体新闻内容及其关注度随时间而变化的特点。
[Purpose/significance] This paper conducted a LDA topic analysis on "the Belt and Road" related news content in official medias and built a basic framework of news topic analysis using LDA model to help the public understand the dynamics and progress of the initiative and its focus in different periods.[Method/process] This paper selected "the Belt and Road" related news on the Chinese government Website during 2015 to 2017, and conducted the topic extraction and heat evolution analysis using LDA model.[Result/conclusion] A total of 30 topics were extracted and summarized as seven categories called policy coordination, facilities connectivity, unimpeded trade, financial integration, people-to-people bond, economic impact and government work. Among them, the policy coordination category has the highest heat during whole time period. Unimpeded trade category and economic impact category are the second and third highest. The heat of some topics, such as "reform and transformation", decline over time, while others like "import and export" increase. These results reflect the changes in the attention of the official media to different news topics related with "the Belt and Road".
[1] 杜德斌, 马亚华. "一带一路":中华民族复兴的地缘大战略[J]. 地理研究, 2015, 34(6):1005-1014.
[2] 姚玉娇. 《人民日报》"一带一路"专题报道新闻框架研究[D].乌鲁木齐:新疆大学, 2017.
[3] 曾润喜, 魏冯. "一带一路"国家战略的舆论引导评价研究[J]. 情报杂志, 2017, 36(5):90-94.
[4] 汪海藻. 《中国日报》与《中国日报·美国版》"一带一路"报道比较研究[D].广州:广东外语外贸大学,2017.
[5] 田作宇. 基于语料库的印度英文报纸中"一带一路"相关新闻的态度研究[D]. 北京:北京外国语大学, 2017.
[6] SALTON G, YANG C S. On the specification of term values in automatic indexing[J]. Journal of documentation, 1973, 29(4):351-372.
[7] ALLAN J, PAPKA R, LAVRENKO V. On-line new event detection and tracking[C]//ACM SIGIR Forum. Amherst:University of Massachusetts, 1998:37-45.
[8] 林南. 基于Web舆情的话题识别与追踪技术研究[D]. 福州:福州大学, 2014.
[9] 陈龙. 新闻热点话题发现及演化分析研究与应用[D]. 南京:南京理工大学, 2017.
[10] LAVRENKO V, ALLAN J, DEGUZMAN E, et al. Relevance models for topic detection and tracking[C]//Proceedings of the second international conference on human language technology research. San Francisco:Morgan Kaufmann Publishers Inc., 2002:115-121.
[11] ZHAI C, LAFFERTY J. A study of smoothing methods for language models applied to ad hoc information retrieval[C]//Proceedings of the 24th annual international ACM SIGIR conference on research and development in information retrieval.New York:ACM,2001:334-342.
[12] HOFMANN T. Probabilistic latent semantic analysis[C]//Proceedings of the fifteenth conference on uncertainty in artificial intelligence. San Francisco:Morgan Kaufmann Publishers Inc., 1999:289-296.
[13] BLEI D M, NG A Y, JORDAN M I. Latent dirichlet allocation[J]. Journal of machine learning research, 2003, 3:993-1022.
[14] BLEI D M, LAFFERTY J D. Dynamic topic models[C]//Proceedings of the 23rd international conference on machine learning. New York:ACM, 2006:113-120.
[15] ALSUMAIT L, BARBARÁ D, DOMENICONI C. On-line LDA:adaptive topic models for mining text streams with applications to topic detection and tracking[C]//Proceedings of the 8th IEEE international conference on data mining. Washington:IEEE Computer Society, 2008:3-12.
[16] 楚克明, 李芳. 基于LDA模型的新闻话题的演化[J]. 计算机应用与软件, 2011, 28(4):4-7.
[17] GRIFFITHS T L, STEYVERS M. Finding scientific topics[J]. Proc. national academy of sciences, 2004, 101(1):5228-5235.
[18] 周振宇. 基于LDA的微博与传统媒体的话题对比研究[D]. 上海:上海交通大学, 2013.
[19] STEVENS K, KEGELMEYER P, ANDRZEJEWSKI D, et al. Exploring topic coherence over many models and many topics[C]//Proceedings of the 2012 joint conference on empirical methods in natural language processing and computational natural language learning. Jeju island:Association for Computational Linguistics, 2012:952-961.
