Abstract
We developed a Hidden Markov Model (HMM) that automatically generates short poem. The HMM was trained using the forward-backward algorithm also known as Baum Welch algorithm. The training process was exhausted by a hundreds of iterations through recursion method. Then we used the Viterbi algorithm to decode all the best possible hidden states to predict the next word, and from the previous predicted word, it will generate another word, then another word until it reaches the desire word length that was set in the program. Afterwards, the model was evaluated using several kinds of readability metrics index which measure the reading difficulty and comprehensiveness of the generated poem. Then, we performed a Turing Test, which participated by 75 college students, who are well versed in poetry. They determined if the generated poems was created by a human or a machine. Based from the evaluation results, the highest readability score index of the generated short poem is in the grade 16th level. While 69.2% of the participants in the Turing Test, agreed that most of the machine generated poems were likely created by some well-known poets and writers.
Highlights
Hidden Markov Model (HMM) has been successfully explored and applied in the fields of medical technology, military, forensics, bioinformatics, data security and even in arts and literatures
Hidden Markov Model (HMM) has become the based algorithm for creating a text generator, text summarizer, lyrics and music generator [1], [17] These applications is belong into one specific area of natural language processing called computational creativity, where the goal of the artificial intelligence (AI) is to change the nature of creative processes, where the machine will compete with human creativeness in terms of writing through the use of different mathematical models and algorithms.[2],[3] One specific product of this area is called “Poetry Generation.”
Most of the published papers focus on the inner performance of the HMM model and used F1-score, precision and recall to calculate its accuracy
Summary
Hidden Markov Model (HMM) has been successfully explored and applied in the fields of medical technology, military, forensics, bioinformatics, data security and even in arts and literatures. HMM models had been widely used to create various types of applications software such as speech recognition, image and signal processing, and even text and poetry generation. Where the machine will automatically generate poem(s) based from historical data or corpus data used to train the AI model. This complex task required a considerable amount of input knowledge (e.g. phonetics, syntax, semantics, grammar and rhymes). Only few had considered to measure the outer performance of the HMM model and evaluated the content features of its generated output. We tested the learning ability of the hidden markov based on the number of iterations performed before it produces a high quality machine generated poem. Where the participants are asked to determine, whether the generated poem was created by machine or human?
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
