The student Mohammed Abdul Jaleel Shnein received his master's degree from the Department of Computer Sciences / University of Technology for his thesis (classification of text using a sophisticated intelligent system). In this thesis, these problems are solved using the proposed classification system based on the logic of fog and genetic algorithm. The proposed system passes through four stages: data collection, pre-processing, extraction and grading. At the data collection stage, this system is based on the Sandy Hurricane event, where data is collected for the time period from 10.27.2012 to 11.7.2012. A set of 1002 tweets taken from the primary data are used as test data and a set of 1000 jitter used as training data. The initial processing stage, which is used to improve the text for the best possible results, where the text contains symbols and mark Hashtak and numbers and words are undesirable. In the initial processing phase, six steps are used, such as handling the Hashtak tag, removing the additions, splitting the text into words, removing the stop words, deriving the words to extract the word origin and determining the correct meaning of the word in the sentence. In the feature extraction phase, eleven features are extracted from each tweet to give a more accurate result. In the classification phase, the proposed system used the fuzzy logic and the genetic algorithm, where the proposed system in the classification phase is going through three steps, namely, inhibition, inference and elimination of the degradation. The Fuzzification phase is used to convert real inputs into misty groups and determine the degree of membership for each value in the vector of features. The Genetic Algorithm is used in the induction step to generate new membership grades. In the Inference phase, a set of ambiguous rules is adopted, where the blurry rules are a set of linguistic values for each feature and contain a classification decision. This was on Thursday 10/1/2019 and on the discussion room in the building of the department. The discussion committee was composed of: Dr. Suhad Mal Allah Kazem and Dr. Hassanein Samir Abdullah from the University of Technology / Computer Sciences Department, Dr. Haitham Abdul Latif Omar from Al-Nahrain University Computer Science Department with the presence of the student supervisors (Dr. Yusra Hussein Ali and Dr. Noha Jamil Ibrahim).
|
Source: Department's Media |
