Text Classification Using Convolutional Neural Networks

Abstract

Most of the information (more than 80%) is stored as text, and text mining is a very important process as it is an initial step in the process of text classification, and this is especially the case in the Arabic language. The Aim of The Study is to classify Arabic texts according to specific categories using advanced performance indicators We used Data Templates as a platform for managing and organizing Apache Spark to solve big data challenges. Apache Spark offers several integrated language APIs. nlp lib was used for text processing. The data is pre-processed through several steps, namely separating the words into one text on the basis of the space between words, cleaning the text of unwanted words, restoring the words to their roots, as well as the feature selection process is a critical step. in text classification. It is a preprocessing technology. In this paper, one way to determine which TF attributes are used how often each feature appears in the document is that they consider the first level of the feature selection process. Then we use TF-IDF to determine the significance of the feature in the document, and this is the last step in the preprocessing Outcomes Text classification . Results were evaluated using advanced performance indicators such as accuracy, Precision and recall. A high accuracy of 96.94% was achieved.The main objective of this paper is to classify basic texts quickly and accurately, according to the results as long as the feature size is suitable, the most advanced technology is superior to other pass rate methods due to the reasonable reliability and perfect pruning level.

Logistics Information Traceability Mechanism of Fresh E-Commerce Based on Image Recognition Technology

“…CNN collects information for the image recognition process that reduces the complexity rate in the recognition process. E-commerce platforms commonly use the backpropagation (BP) algorithm for the logistics image recognition process [ 9 , 10 ]. BP algorithm predicts the actual information and patterns about an image and finds out the appropriate set of data for further analysis.…”

English Text Similarity Detection Algorithm Based on Animal Algorithm

“…By filtering and refining the vocabulary of the document to obtain a keyword set, and then organizing the keywords to obtain a unit vector in a multidimensional space, the unit direction of each dimension represents the information of a keyword in the text, and the direction of extension in this dimension Determined by related weights. The formula is expressed as formula (1), where V(f) represents the space vector of text f, and wn represents the feature vector representation of the nth keyword. In the database, the vector dimension n of the model should take the same value.…”

“…Text mining involves text modeling and representation, text word weight calculation, text similarity calculation, text information extraction, application research of potential information in text, etc [1,2]. Among them, the text similarity algorithm is a crucial one in text mining Algorithms are the link between basic research such as text modeling and representation and upper-level application research of text latent information [3].…”