Research on Automatic Summary Method for Futures Research Reports Based on TextRank

Futures research reports, authored by futures analysts, delve into various aspects such as futures contracts, the macro-environment, industry trends, and industry chains. These reports provide crucial information for comprehending trends in futures trading. The accurate identification and extraction of key data from extensive futures research reports poses a significant challenge for market regulators. An automatic summarization method for futures research reports is proposed. Firstly, key sentences are extracted from a futures research report using a self-constructed word list integrated lexical chain in the field of futures. Secondly, key sentences are represented as vectors that incorporate term frequency-inverse document frequency. Thirdly, the TextRank algorithm is employed to extract candidate summary. Finally, the maximum marginal relevance algorithm is employed to supplement some crucial but easily missed key sentences, thereby ensuring the informativeness of the summary. ROUGE is used to verify the validity of the method. Experimental results demonstrate that our method is both simple and effective in automatic summarization of futures research reports.