Chronological Ordering Based on Context Overlap Detection

2. Related Works

The goal is to determine a most probable replacement of sentences or, in other words, reconstruct speak structure of sentences gathered from multiple sources. When a human is asked to make an arrangement of sentences, he or she may perform this task without difficulty just as writing out thoughts in a text. However, they considered what accomplishes this task since computers are unaware of order of things by nature. There are many searches in ordering sentences techniques. Tiedan Zhu et al. (2012) presented an improved approach to sentence ordering in multi document summarization. They rely on using sentences logical closeness by defining the notation a→b to represent that a and b are adjacent. The ‘adjacent’ here means: a and b are coherent enough to be connected together in a document (summary); second, a precedes b. then they defined the sentence-chain which is a chain of adjacent sentences, i.e. A = (a1→a2→…→an-1→an) is a sentence-chain with the length of n, where ai is a sentence. Then Used the arrow with sentence-chains, A→B= (a1→a2→…→an-1→an→b1→b2→…→bm-1→bm). The result of two sentence-chains connected with an arrow is still a sentence-chain.