In earlier study of news texts, transitivity analysis was usually confined to the comparison between the headlines or single clauses of two pieces of selected news. 在早期的新聞?wù)Z(yǔ)篇研究中，及物性分析往往局限于對兩篇新聞的標題和個(gè)別短句進(jìn)行比較分析。
We propose an algorithm based on spectral analysis on graph Laplacian to decompose the problem and perform parameter relativity analysis to provide state abstraction. 這篇研究基于針對拉普拉斯圖的光譜分析提出新的問(wèn)題切割演算法，并在切割過(guò)的子問(wèn)題上分析參數相關(guān)度來(lái)施行狀態(tài)抽象化。
