Abstract Sentiment analysis is the process to study of people opinion, emotion and way of considering a matter and take the decision into different categorize like positive, negative and neutral in data mining. The sentiment analysis is used to find out negation within the text using Natural Language Processing rules. Our aim is to detect negation affect on consumer reviews which... Sentiment analysis is the process to study of people opinion, emotion and way of considering a matter and take the decision into different categorize like positive, negative and neutral in data mining. The sentiment analysis is used to find out negation within the text using Natural Language Processing rules. Our aim is to detect negation affect on consumer reviews which look like positive but exactly negative in sense. A number of different approaches have been used, but these approaches do not provide an efficient and appropriate way of calculating negation sense in sentiment analysis. The proposed modified negation approach presents a way of calculating negation identification and is helpful to improve classification accuracy. The main achievement of this approach is that it is helpful for calculating the negation in sentiment analysis without the words not, no, n’t, never etc. This method produced a significant result for review classification by accuracy, precision, and recall. Read More Read Less

Abstract Pictures consisting of computer characters are called ASCII art. They are widely used in computer texts because they enrich expression in texts. To deal with texts including ASCII art objects, ASCII art extraction methods, which detect ASCII art objects in a given text, are now demanded. For example, such methods are needed to find ASCII art objects in texts for... Pictures consisting of computer characters are called ASCII art. They are widely used in computer texts because they enrich expression in texts. To deal with texts including ASCII art objects, ASCII art extraction methods, which detect ASCII art objects in a given text, are now demanded. For example, such methods are needed to find ASCII art objects in texts for sentiment analysis based on ASCII art objects. Such methods are also needed to remove ASCII art objects from texts in preprocess for natural language processing because non-verbal texts of ASCII art makes texts noisy. Our research group and another research group independently proposed two different ASCII art extraction methods, which are a run-length encoding based method and a byte pattern based method respectively. Both of the methods use ASCII art recognizers constructed by machine learning algorithms, but they use different attributes of texts. In this paper, we compare the two methods by ASCII art extraction experiments where training texts and testing texts are in English and Japanese. According to our experimental results, the two methods are competitive if training texts and testing texts are in a same set of languages, but the run-length encoding based method works better than the byte pattern based method if training texts and testing texts are in different sets of languages. Read More Read Less

Abstract Since twenty-first century, the increase in communication among different countries has made the need for the language translation of the enterprises and individuals more and more. Artificial translation is accurate, but the cost is too high and time-consuming; while the cost of the machine translation is not only low, but the speed is fast. However, the accuracy of machine translation... Since twenty-first century, the increase in communication among different countries has made the need for the language translation of the enterprises and individuals more and more. Artificial translation is accurate, but the cost is too high and time-consuming; while the cost of the machine translation is not only low, but the speed is fast. However, the accuracy of machine translation has been criticized by users, therefore, how to build a new generation of machine translation system to improve the accuracy has been imminent. Based on this status, a reliable English-Chinese machine translation system based on artificial intelligence is established in this paper, and the principles that should be followed in the process of establishing the system are described in detail. The overall framework, the translation algorithm and the working flow of the system are discussed, and the sentence alignment method based on the translation is proposed. The research results show that the reliable English-Chinese machine translation system based on artificial intelligence designed in this paper can improve the credibility and accuracy of machine translation. Read More Read Less