SRFW: a simple, fast and effective text classification algorithm

Bioinspired design is the adaptation of methods, strategies, or principles found in nature to solve engineering problems. One formalized approach to bioinspired solution seeking is the abstraction of the engineering problem into a functional need and then seeking solutions to this function using a keyword type search method on text based biological knowledge. These function keyword search approaches have shown potential for success, but as with many text based search methods, they produce a large number of results, many of little relevance to the problem in question. In this paper, we develop a method to train a computer to identify text passages more likely to suggest a solution to a human designer. The work presented examines the possibility of filtering biological keyword search results by using text mining algorithms to automatically identify which results are likely to be useful to a designer. The text mining algorithms are trained on a pair of surveys administered to human subjects to empirically identify a large number of sentences that are, or are not, helpful for idea generation. We develop and evaluate three text classification algorithms, namely, a Naïve Bayes (NB) classifier, a k nearest neighbors (kNN) classifier, and a support vector machine (SVM) classifier. Of these methods, the NB classifier generally had the best performance. Based on the analysis of 60 word stems, a NB classifier's precision is 0.87, recall is 0.52, and F score is 0.65. We find that word stem features that describe a physical action or process are correlated with helpful sentences. Similarly, we find biological jargon feature words are correlated with unhelpful sentences.

With the rapid development of World Wide Web, text categorization has become the key technology in organizing and processing large volume of document data. There are a variety of text categorization methods such as k nearest neighbor (kNN) and support vector machine (SVM). However, those methods are either too complicated or not effective enough. In this paper, we present a new method called relative importance in category (RIIC), which is simpler than most methods and has a lower time complexity. To verify the performance of RIIC, we build a text categorization system (TCS) based on RIIC and compare our system with TCS based on kNN and SVM. Experimental results show that in most cases the performance of RIIC is better than kNN and SVM.

In text classification community, k-nearest neighbor (kNN) and support vector machine (SVM) are all effective classifiers, but both of them have their own drawbacks. kNN involves high cost to classify a new document when training set is large; SVM is too sensitive to the noisy data when the noisy data is close to the hyperplane it suffers. So one hybrid algorithm based on variable precision rough set is proposed. It combines the strength of both KNN and SVM techniques and overcomes their weaknesses. Finally some experiments are carried out to compare the efficiency and classification accuracy with different classification algorithms. Results show that the proposed method achieves significant performance improvement.

Nearest neighbour classifier and support vector machine (SVM) are successful classifiers that are widely used in many important application areas. But both these classifiers suffer from the curse of dimensionality. Nearest neighbour search, in high dimensional data, using Euclidean distance is questionable since all the pair wise distances seem to be almost the same. In order to overcome this problem, we propose a novel classification system based on majority voting. Firstly, we partition the features into a number of blocks and construct a classifier for each block. The majority voting is then performed across all classifiers to determine the final class label. Classification is also performed using non-negative matrix factorisation (NNMF) that embeds high dimensional data into low dimensional space. Experiments were conducted on three of the benchmark datasets and the results obtained showed that the proposed system outperformed the conventional classification using both k-nearest neighbour (k-NN) and support vector machine (SVM) classifiers. The proposed system also showed better performance when compared with the classification performance of 1NN and SVM classifier using NNMF-based dimensionally reduced data.

The advent of internet has lead to colossal development of e-learning frameworks. The efficiency of such systems however relies on the effectiveness and fast content based retrieval approaches. This paper presents a methodology for efficient search and retrieval of lecture videos based on Machine Learning (ML) text classification algorithm. The text transcript is generated exclusively from the audio content extracted from the video lectures. This content is utilized for the summary and keyword extraction which is used for training the ML text classification model. An optimized search is achieved based on the trained ML model. The performance of the system is compared by training the system using Naive Bayes, Support Vector Machine and Logistic Regression algorithms. Performance evaluation was done by precision, recall, F-score and accuracy of the search for each of the classifiers. It is observed that the system trained on Naive Bayes classification algorithm achieved better performance both in terms of time and also with respect to relevancy of the search results.

Natural Language Processing (NLP) is a science that integrates computer knowledge, mathematical knowledge and linguistic knowledge, and text classification and recognition is considered to be an important research field and direction of natural language processing. This paper mainly studies the realization of text classification model through the processing method of text data in natural language processing and the theoretical knowledge and technical means of machine learning. We summarize the existing text classification algorithms, analyze their applicable scenarios, and optimize on the basis of these algorithm models. The paper proposes a Chinese text classification algorithm based on weight preprocessing. Algorithm based on the weight preprocessing link, so that the optimized classifier model can improve the accuracy of the existing text classifier. In this paper, the English Newsgroups corpus is used for experimental verification. The experimental results show that the classification accuracy and accuracy of the improved algorithm are better than those of the traditional text classification algorithm, thereby improving the accuracy of English text classification.

Machine learning methods are widely used in automated technologies. Classification prediction is a machine learning based on data mining. Today, many technological devices can make new predictions by gaining experience from past data with machine learning methods. Machine learning is widely studied in two types, supervised and unsupervised. The limits of the objectives in supervised learning are predetermined. In unsupervised learning, there are no predetermined targets. In this learning, the machines are required to determine the targets automatically. Prediction process is one of the basic components of machine learning. Machines need to use some algorithms in order to perform the prediction process on the basis of data mining. k nearest neighbor (kNN), Naive Bayes (NB), Decision Tree (DT) and Support Vector Machine (SVM) algorithms are used mostly. k nearest neighbor (kNN), Naive Bayes (NB), Decision Tree (DT) and Support Vector Machine (SVM) algorithms are used mostly. These algorithms can be applied with the help of some tools on data sets. In this study, kNN algorithm was used to estimate Iris data set classification using Orange tool. The success of the KNN algorithm depends on using the correct attribute and changing the optimum k value. As a result of the tests, it was determined that when the k neighbor value was selected as 15, it was the most suitable k neighbor value, providing 98.67% classification prediction success in the Iris dataset.

The decision to raise the Value Added Tax (VAT) rate to 12% in 2025 has elicited diverse public responses, especially on the social media platform X (previously Twitter). This study seeks to categorize public attitudes regarding the policy utilizing two text classification algorithms: Support Vector Machine (SVM) and Multinomial Naïve Bayes. The Synthetic Minority Over-sampling Technique (SMOTE) was applied to the training data to address class imbalance in the sentiment data (positive, negative, neutral). Data were gathered via web crawling using terms relevant to the VAT rise issue, then underwent preprocessing, TF-IDF transformation, and division into two scenarios (80:20 and 70:30). The models' performance was assessed using accuracy, precision, recall, F1-score, and Area Under the Curve (AUC) metrics. The findings indicate that the SVM method without SMOTE achieved optimal performance, with an accuracy of 90.52% and an AUC of 0.860. Simultaneously, the Naïve Bayes algorithm showed improved performance after SMOTE, especially in recall and F1-score, while its overall performance remained inferior to that of SVM. Consequently, SVM is considered more reliable for sentiment classification tasks, although SMOTE has demonstrated efficacy in improving Naïve Bayes performance, particularly for addressing imbalanced datasets.

Semantic text classification: A survey of past and recent advances

Introduction: Since the delay or mistake in the diagnosis of mood disorders due to the similarity of their symptoms hinders effective treatment, this study aimed to accurately diagnose mood disorders including psychosis, autism, personality disorder, bipolar, depression, and schizophrenia, through modeling and analyzing patients' data . Method : Data collected in this applied developmental research included 996 records with 130 features obtained by interviewing and completing questionnaires in a mental hospital in the city of Sari, Iran in 2021. After preprocessing, the number of features was reduced to 91, and then through Principal Component Analysis (PCA) reduced to 35 factors. Modeling was done in Python software with K -Nearest Neighbor (KNN), Naive Bayes (NB), Decision Tree (DT), Random Forest (RF), Logistic Regression (LR), and Support Vector Machine (SVM) algorithms. The models were evaluated to select algorithms with higher accuracy. Particle Swarm Optimization (PSO) and Genetic Algorithm (GA) were applied to determine the optimal parameters of the selected algorithms . Results: Among the machine learning algorithms, random forest with 91% accuracy and support vector machine with 90% accuracy showed better performance. The genetic algorithms did not make any notable increase in prediction accuracy. Whereas considering N=30, T=150, W=0.9, c1=2, and c2=2 in the particle swarm optimization algorithm increased the prediction accuracy up to 3.3 %. Conclusion : With less classification error compared to similar studies, the PSO -SVM model designed in this study can be used in patient data monitoring with acceptable accuracy and can be used in intelligent systems in psychiatric centers.

Big multimedia data is heterogeneous in essence, that is, the data may be a mixture of video, audio, text, and images. This is due to the prevalence of novel applications in recent years, such as social media, video sharing, and location based services (LBS), etc. In many multimedia applications, for example, video/image tagging and multimedia recommendation, text classification techniques have been used extensively to facilitate multimedia data processing. In this paper, we give a comprehensive review on feature selection techniques for text classification. We begin by introducing some popular representation schemes for documents, and similarity measures used in text classification. Then, we review the most popular text classifiers, including Nearest Neighbor (NN) method, Naive Bayes (NB), Support Vector Machine (SVM), Decision Tree (DT), and Neural Networks. Next, we survey four feature selection models, namely the filter, wrapper, embedded and hybrid, discussing pros and cons of the state-of-the-art feature selection approaches. Finally, we conclude the paper and give a brief introduction to some interesting feature selection work that does not belong to the four models.

In text classification, feature selection is essential to improve the classification effectiveness. This paper provides an empirical study of a feature selection method based on genetic algorithms for different text representation methods. This feature selection algorithm can accomplish two goals: in one hand is the search of a feature subset such that the performance of classifier is best; in other hands is find a feature subset with the smallest dimensionality which achieves higher accuracy in classification. To evaluate the performance of this approach, three from the best classifiers have been selected: Naive Bayes (NB), Nearest Neighbors (KNN) and Support Vector Machines (SVMs). Our objective is to determine whether the genetic algorithms based feature selection will improve the performances in text classification with smaller size using F-measure. Experimentations were carried out on two benchmark document collections 20Newsgroups, and Reuters-21578. And the results were very interesting.

Applications of Deep Belief Nets (DBN) to various problems have been the subject of a number of recent studies ranging from image classification and speech recognition to audio classification. In this study we apply DBNs to a natural language understanding problem. The recent surge of activity in this area was largely spurred by the development of a greedy layer-wise pretraining method that uses an efficient learning algorithm called Contrastive Divergence (CD). CD allows DBNs to learn a multi-layer generative model from unlabeled data and the features discovered by this model are then used to initialize a feed-forward neural network which is fine-tuned with backpropagation. We compare a DBN-initialized neural network to three widely used text classification algorithms: Support Vector Machines (SVM), boosting and Maximum Entropy (MaxEnt). The plain DBN-based model gives a call-routing classification accuracy that is equal to the best of the other models. However, using additional unlabeled data for DBN pre-training and combining DBN-based learned features with the original features provides significant gains over SVMs, which, in turn, performed better than both MaxEnt and Boosting.

The Support Vector Machine (SVM) is an effective classification tool. Though extremely effective, SVMs are not a panacea. SVM training and testing is computationally expensive. Also, tuning the kernel parameters is a complicated procedure. On the other hand, the Nearest Neighbor (KNN) classifier is computationally efficient. In order to achieve the classification efficiency of an SVM and the computational efficiency of a KNN classifier, it has been shown previously that, rather than training a single global SVM, a separate SVM can be trained for the neighbourhood of each query point. In this work, we have extended this Local SVM (LSVM) formulation. Our Local Adaptive SVM (LASVM) formulation trains a local SVM in a modified neighborhood space of a query point. The main contributions of the paper are twofold: First, we present a novel LASVM algorithm to train a local SVM. Second, we discuss in detail the motivations behind the LSVM and LASVM formulations and its possible impacts on tuning the kernel parameters of an SVM. We found that training an SVM in a local adaptive neighborhood can result in significant classification performance gain. Experiments have been conducted on a selection of the UCIML, face, object, and digit databases.

Brain Computer Interface (BCI) technology is used to help patients who do not have control over motor neurons such as ALS or paralyzed patients, to communicate with outer world. This work aims to classify motor imageries using real-time EEG dataset, which was published by Graz University, Austria. The dataset consists of two-channel EEG signals of right-hand movement imagery and left-hand movement imagery of 8 subjects. There are a total of 120 motor imagery trials (60 left and 60 right) EEG signals recorded from each subject. EEG signals are filtered and feature vectors were extracted that consist of 24, 32 and 40 relative band power values (RBPV). In this work, feature vectors classified by three different methods, linear discriminant analysis (LDA), K nearest neighbor (KNN) and support vector machines (SVM). Results show that best performance was achieved by 24 RBPV feature vector and LDA classification method.