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. 在早期的新聞語篇研究中，及物性分析往往局限於對兩篇新聞的標題和個別短句進行比較分析。
We propose an algorithm based on spectral analysis on graph Laplacian to decompose the problem and perform parameter relativity analysis to provide state abstraction. 這篇研究基於針對拉普拉斯圖的光譜分析提出新的問題切割演演算法，並在切割過的子問題上分析參數相關度來施行狀態抽象化。
