No abstract

Building systems with capability of natural language understanding (NLU) has been one of the oldest areas of AI. An essential component of NLU is to detect logical succession of events contained in a text. The task of sentence ordering is proposed to learn succession of events with applications in AI tasks. The performance of previous works employing statistical methods is poor, while the neural networks-based approaches are in serious need of large corpora for model learning. In this paper, we propose a method for sentence ordering which does not need a training phase and consequently a large corpus for learning. To this end, we generate sentence embedding using BERT pre-trained model and measure sentence similarity using cosine similarity score. We suggest this score as an indicator of sequential events' level of coherence. We finally sort the sentences through brute-force search to maximize overall similarities of the sequenced sentences. Our proposed method outperformed other baselines on ROCStories, a corpus of 5-sentence humanmade stories. The method is specifically more efficient than neural network-based methods when no huge corpus is available. Among other advantages of this method are its interpretability and needlessness to linguistic knowledge.

No abstract

Segmentation approaches, as processes that divide word into smaller parts which contain one letter at most, have important effect on cursive word recognition. While online cursive word recognition became applied technology in Latin and Chinese languages, complex structural features in Arabic-based script made it an important field of study in Persian and Arabic languages. In this paper, by introducing of Standard Persian Handwriting, we proposed a novel approach to segmentation online Persian cursive script based on width of letter's body in Persian language. Results are shown 99.86% accuracy in detection of expected segmentation points, while recognized extra points reduced 93.73% compared to our previous methods.

No abstract