A Comparative Performance Study of Classification Models for Opinion Mining

Sentiment Analysis is a widely used text classification technique. It breaks down any given text or comments and classify the text either as positive or negative based on the views conveyed in it. Previous works done on sentiment classification used either lexicon based approach or machine learning techniques. Likewise, the major drawback of the existing systems was the focus on only binary classification of review such as positive or negative. Ignorance of the neutral review will result in misinterpretation of a customer's opinion about a product or movie, which will degrade the business or trend. In case of using only lexicon based approach, the system highly depends on the selection of lexicon resource and dictionary. In case of system built only using machine learning approach, the performance of the system depends on the algorithms chosen. This work presents a hybrid model to resolve the neutral class too. The proposed work combines a lexical approach (SentiWordNet) with the machine learning algorithms such as Support Vector Machine, Decision Tree, Logistic Regression and Naive Bayes for sentiment analysis to resolve the neutral opinions beyond the binary categorization of the customer's review. We have also compared the performance of these four machine learning algorithms along with the lexicon approach. The results proved that Support Vector Machine and Logistic Regression algorithms outperform the other two algorithms with an accuracy of about 80% which is on average differs by 6% to 10% when compared to other algorithms.

The rapid dissemination of information related to the K-Pop world, facilitated by social media, has made it easier to follow developments and controversies. One notable case that sparked extensive discussion on Twitter was the bullying allegations against Kim Garam of LE SSERAFIM. Researchers, using Twitter data, sought to analyze Indonesian public sentiment regarding this case through sentiment analysis, which classifies opinions as positive or negative. For processing textual data, text mining methods, particularly classification techniques, are employed. Two popular algorithms in text mining are the Naive Bayes classifier and the support vector machine (SVM). The Naive Bayes classifier is favored for its speed, simplicity, and high accuracy, while the SVM excels at identifying a hyperplane that maximizes the margin between classes. In this study, sentiment classification results were labeled as either positive or negative. The comparison between the Naive Bayes classifier and SVM for classifying responses to Kim Garam's bullying case on Twitter showed high accuracy rates: 93% for Naive Bayes and 97% for SVM. The higher accuracy of the SVM algorithm indicates its superiority over the Naive Bayes classifier in this context.

Text mining is a special case of data mining which explore unstructured or semi-structured text documents, to establish valuable patterns and rules that indicate trends and significant features about specific topics. Text mining has been in pattern recognition, predictive studies, sentiment analysis and statistical theories in many areas of research, medicine, financial analysis, social life analysis, and business intelligence. Text mining uses concept of natural language processing and machine learning. Machine learning algorithms have been used and reported to give great results, but their performance of machine learning algorithms is affected by factors such as dataset domain, number of classes, length of the corpus, and feature selection techniques used. Redundant attribute affects the performance of the classification algorithm, but this can be reduced by using different feature selection techniques and dimensionality reduction techniques. Feature selection is a data preprocessing step that chooses a subset of input variable while eliminating features with little or no predictive information. Feature selection techniques are Information gain, Term Frequency, Term Frequency-Inverse document frequency, Mutual Information, and Chi-Square, which can use a filters, wrappers, or embedded approaches. To get the most value from machine learning, pairing the best algorithms with the right tools and processes is necessary. Little research has been done on the effect of feature selection techniques on classification accuracy for pairing of these algorithms with the best feature selection techniques for optimal results. In this research, a text classification experiment was conducted using incident management dataset, where incidents were classified into their resolver groups. Support vector machine (SVM), K-Nearest Neighbors (KNN), Naïve Bayes (NB) and Decision tree (DT) machine learning algorithms were examined. Filtering approach was used on the feature selection techniques, with different ranking indices applied for optimal feature set and classification accuracy results analyzed. The classification accuracy results obtained using TF were, 88% for SVM, 70% for NB, 79% for Decision tree, and KNN had 55%, while Boolean registered 90%, 83%, 82% and 75%, for SVM, NB, DT, and KNN respectively. TF-IDF, had 91%, 83%, 76%, and 56% for SVM, NB, DT, and KNN respectively. The results showed that algorithm performance is affected by feature selection technique applied. SVM performed best, followed by DT, KNN and finally NB. In conclusion, presence of noisy data leads to poor learning performance and increases the computational time. The classifiers performed differently depending on the feature selection technique applied. For optimal results, the classifier that performed best together with the feature selection technique with the best feature subset should be applied for all types of data for accurate classification performance. Keywords: Text Classification, Supervised Machine Learning, Feature Selection DOI : 10.7176/JIEA/9-3-06 Publication date :May 31 st 2019

Objective: Using methylation characteristics of human genes to construct machine learning predictive models for screening cervical cancer and precancerous lesions. Methods: Human DNA methylation detection was performed on 224 cervical exfoliated cell specimens from the Cancer Hospital of the Chinese Academy of Medical Sciences, Tianjin Central Hospital of Gynecology Obstetrics, Xinmi Maternal and Child Health Hospital of Henan Province, West China Second Affiliated Hospital of Sichuan University, and Heping Hospital Affiliated to Changzhi Medical College collected during April 2014 and March 2015. The hypermethylated gene fragments related to cervical cancer were selected by high-density, high-association, and hypermethylated gene fragment screening and the LASSO regression algorithm. Taking cervical intraepithelial neoplasia grade 2 (CIN2) or more severe lesions as the research outcome, machine learning predictive models based on the random forest (RF), naive Bayes (NB), and support vector machine (SVM) algorithm, respectively, were constructed. A total of 144 outpatient specimens were used as the training set and 80 cervical exfoliated cell specimens from women participating in the cervical cancer screening program were used as the test set to verify the predictive models. Using histological diagnosis results as the gold standard, the detection efficacy for CIN2 or more severe lesions of the three machine learning predictive models were compared with that of the human papilloma virus (HPV) detection and cytological diagnosis. Results: In the training set of 144 cases, there were 34 cases of HPV positivity, with a positive rate of 23.61%. Cytologically, there were 37 cases diagnosed as no intraepithelial lesion or malignancy (NILM), and 107 cases diagnosed as atypical squamous cells of undetermined significance (ASC-US) or above. Histologically, there were 28 cases without cervical intraepithelial neoplasia or benign cervical lesions, 31 cases of CIN1, 18 cases of CIN2, 31 cases of CIN3, and 36 cases of squamous cell carcinoma. Seven hypermethylated gene fragments were selected from 45 genes, and three machine learning prediction models based on the RF, NB, and SVM algorithm, respectively, were constructed. In the validation set of 80 cases, there were 28 cases of HPV positivity, with a positive rate of 35.00%. Cytologically, there were 65 cases diagnosed as NILM and 15 cases as ASC-US or above. Histologically, there were 39 cases without cervical intraepithelial neoplasia or benign cervical lesions, 10 cases of CIN1, 10 cases of CIN2, 11 cases of CIN3, and 10 cases of squamous cell carcinoma. In the validation set, the area under the curve (AUC) values of the RF model, NB model, SVM model, HPV detection, and cytological diagnosis of CIN2 or above were 0.90, 0.88, 0.82, 0.68, and 0.45, respectively. The DeLong test showed that there was no statistically significant difference in the AUC values between the RF, NB, and SVM models (all P＞0.05), and the AUC values of the RF and NB models were higher than that of HPV detection (both P＜0.01), and the AUC values of the RF, NB, and SVM models were higher than that of cytological diagnosis (all P＜0.01). Compared with the NB model, the sensitivity of the RF model was similar (80.65% vs. 77.42%), but the specificity of the NB model was much higher than that of the RF model (93.88% vs. 73.47%). Conclusion: Among the machine learning prediction models for cervical cancer and precancerous lesions constructed based on human DNA methylation, the NB model has good predictive performance for CIN2 and above lesions, and may be used for screening of cervical cancer and precancerous lesions.

The use of Machine Learning (ML) and Sentiment Analysis (SA) on data from microblogging sites has become a popular method for stock market prediction. In this work, we developed a model for predicting stock movement utilizing SA on Twitter and StockTwits data. Stock movement and sentiment data were used to evaluate this approach and validate it on Microsoft stock. We gathered tweets from Twitter and StockTwits, as well as financial data from Finance Yahoo. SA was applied to tweets, and seven ML classification models were implemented: K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Logistic Regression (LR), Naïve Bayes (NB), Decision Tree (DT), Random Forest (RF) and Multilayer Perceptron (MLP). The main novelty of this work is that it integrates multiple SA and ML methods, emphasizing the retrieval of extra features from social media (i.e., public sentiment), for improving stock prediction accuracy. The best results were obtained when tweets were analyzed using Valence Aware Dictionary and sEntiment Reasoner (VADER) and SVM. The top F-score was 76.3%, while the top Area Under Curve (AUC) value was 67%.

Mining data over streams in one pass and using constant memory is a challenging task. Decision trees are one of the most popular classifiers for both batch and incremental learning due to their high degree of interpretability, ease of construction and good accuracy. The most popular decision tree for stream classification is Hoeffding Tree based on Hoeffding bound. Literature shows a few variants of decision trees based on different bounds. The default class prediction method adopted in decision tree is “majority class” approach. Later, the accuracy of prediction was scaled up by a hybrid decision tree where Naive Bayes classifier was used for prediction. Kernel Density Estimation (KDE) is employed in Flexible Naive Bayes for classification. However, it is suitable for modeling static data set. This paper proposes an Incremental Flexible Naive Bayes (IFNB) based hybrid decision tree paradigm that uses KDE to model continuous attributes at leaf nodes of the tree for improving the class prediction accuracy. Experimental results on both synthetic and real dataset show that the proposed IFNB based leaf classifiers achieves improvement over the class prediction methods adopted in existing decision trees for data streams.

The field of sentiment mining and text summarization has evoked the interest of many scientists and researchers over the last few years, as the textual data has become useful for many real-world applications and challenges. Sentiment Analysis and Opinion Mining is the most popular field for analyzing and discovering insights from text data from various sources, such as Facebook, Twitter and Amazon, Zomato, etc. It involves a computational study of an individual's behavior in terms of buying interest and then extracting his opinions on the business entity of the company. This entity can be viewed as an event, individual, blog post or product experience. Scholars in the fields of natural language processing, data mining, machine learning and others have tested a variety of methods for automating sentiment analysis. These reviews are increasing on a daily basis, as a result of which the summarization of the reviews plays a role where the text is summarized as needed, which provides useful information from a large number of reviews. It is very difficult for a human being to extract and interpret useful data from a very large file. In the text analysis, the value of sentences is decided on the basis of the linguistic characteristics of sentences. This paper provides a comprehensive review of current and past work on sentiment analysis and text description. In this research work, a new hybrid classification system is proposed based on coupling classification methods using arcing classifiers and their quality is evaluated within terms of accuracy. The Classifier Collection was constructed using Naïve Bayes (NB), Support Vector Machine (SVM) and Genetic Algorithm (GA). The proposed work consists of a comparative study of the efficacy of the ensemble technique for sentiment classification. The feasibility and benefits of the proposed approaches are demonstrated by a restaurant review that is widely used in the field of sentiment classification. A wide range of comparative studies is performed and, ultimately, some in-depth analysis is addressed and conclusions are drawn on the efficacy of the ensemble technique for sentiment classification.

Sentiment Analysis using Contextual Approach for E-Commerce Reviews

Opinion is a statement conveyed by a person or group of people in addressing a problem by providing predictions or expectations about the event. No guarantee that an opinion automatically will be true because it is not reinforced by the facts, it is subjective, and there is a different opinion about an event. Everyone has different views and same rights to express opinions or give opinions toward particular event. Public opinion is view of someone for certain problem comes out due to prior conversation with another person who may have an effect on the opinion given. Public opinion comes from a discussion process in addressing the problem then lead to a conclusion as a joint decision and a shared opinion. One of the media to convey public opinion is through social media like twitter. Public opinion about the election of West Java governor candidate period 2018-2023 on twitter was increasingly widespread. There are several sentiments emerged for four candidates elected on twitter accounts such as Ridwan Kamil-Uu Ruzhanul Ulum, Tubagus Hasanuddin-Anton Charliyan, Sudrajat-Ahmad Syaikhu, and Deddy Mizwar-Dedi Mulyadi. Therefore, it is necessary to classify the sentiments to the existing opinion so that it can be predicted in advance which of the governor candidate pair of West Java who has more positive and predictable sentiments will be elected as governor period 2018-2023. The data used by the researchers is tweet in Indonesian Language with keywords Rindu, Hasanah, Asyik, 2DM with datasets number is 800 tweets. There are many classification techniques commonly used for opinion sentiment analysis. This study compares two classification techniques namely Support Vector Machine Algorithm (SVM) and Naive Bayes Classifier (NBC). The results show that the Algorithm of Naive Bayes Classifier (NBC) has a higher accuracy level of Support Vector Machine (SVM), up to 94% for Deddy Mizwar-Dedi Mulyadi.

Background and aimsThe birth weight of a newborn is a crucial factor that affects their overall health and future well-being. Low birth weight (LBW) is a widespread global issue, which the World Health Organization defines as weighing less than 2,500 g. LBW can have severe negative consequences on an individual’s health, including neonatal mortality and various health concerns throughout their life. To address this problem, this study has been conducted using BDHS 2017–2018 data to uncover important aspects of LBW using a variety of machine learning (ML) approaches and to determine the best feature selection technique and best predictive ML model.MethodsTo pick out the key features, the Boruta algorithm and wrapper method were used. Logistic Regression (LR) used as traditional method and several machine learning classifiers were then used, including, DT (Decision Tree), SVM (Support Vector Machine), NB (Naïve Bayes), RF (Random Forest), XGBoost (eXtreme Gradient Boosting), and AdaBoost (Adaptive Boosting), to determine the best model for predicting LBW. The model’s performance was evaluated based on the specificity, sensitivity, accuracy, F1 score and AUC value.ResultsResult shows, Boruta algorithm identifies eleven significant features including respondent’s age, highest education level, educational attainment, wealth index, age at first birth, weight, height, BMI, age at first sexual intercourse, birth order number, and whether the child is a twin. Incorporating Boruta algorithm’s significant features, the performance of traditional LR and ML methods including DT, SVM, NB, RF, XGBoost, and AB were evaluated where LR, had a specificity, sensitivity, accuracy and F1 score of 0.85, 0.5, 85.15% and 0.915. While the ML methods DT, SVM, NB, RF, XGBoost, and AB model’s respective accuracy values were 85.35%, 85.15%, 84.54%, 81.18%, and 84.41%. Based on the specificity, sensitivity, accuracy, F1 score and AUC, RF (specificity = 0.99, sensitivity = 0.58, accuracy = 85.86%, F1 score = 0.9243, AUC = 0.549) outperformed the other methods. Both the classical (LR) and machine learning (ML) models’ performance has improved dramatically when important characteristics are extracted using the wrapper method. The LR method identified five significant features with a specificity, sensitivity, accuracy and F1 score of 0.87, 0.33, 87.12% and 0.9309. The region, whether the infant is a twin, and cesarean delivery were the three key features discovered by the DT and RF models, which were implemented using the wrapper technique. All three models had the identical F1 score of 0.9318. However, “child is twin” was recognized as a significant feature by the SVM, NB, and AB models, with an F1 score of 0.9315. Ultimately, with an F1 score of 0.9315, the XGBoost model recognized “child is twin” and “age at first sex” as relevant features. Random Forest again beat the other approaches in this instance.ConclusionsThe study reveals Wrapper method as the optimal feature selection technique. The ML method outperforms traditional methods, with Random Forest (RF) being the most effective predictive model for Low-Birth-Weight prediction. The study suggests that policymakers in Bangladesh can mitigate low birth weight newborns by considering identified risk factors.

Machine Learning Techniques for Breast Cancer Prediction

Medical datasets have attracted the research community for possible analysis and suitable prediction, which helps the human to take proper precautions in preventing future diseases. To perform related operations, data mining techniques have been widely used in developing decision support systems for disease prediction through a set of medical datasets. This work proposes a new predictive model for disease prediction using pre-processing techniques for various disease datasets. The proposed model not only analyses the datasets also improves the performance by using ensemble methods. To process the datasets, pre-processing techniques such as discretization, resampling, principal component, and decision tree have been used. To classify the datasets, classification techniques such as Support Vector Machines (SVM), K-Nearest Neighbors (KNN), Naive Bayes (NB), Decision Tree (DT), and Random Forest (RF) have been used. The algorithms are applied with 10 fold validation technique. A predictive analysis has also been performed on various disease datasets, where every dataset results in significant improvement for various performance measures. We perform a predictive analysis on the datasets such as CKD (Chronic Kidney Disease), Cardiovascular Disease (CVD) or heart, Diabetes, Hepatitis disease, Cancer disease and ILPD (Indian Liver Patient disease). Experimental results show that the proposed predictive model outperforms in terms of better accuracy.

Single-cell RNA-seq overcomes the shortcomings of conventional transcriptome sequencing technology and could provide a powerful tool for distinguishing the transcriptome characteristics of various cell types in biological tissues, and comprehensively revealing the heterogeneity of gene expression between cells. Many Intelligent Computing methods have been presented to infer gene regulatory network (GRN) with single-cell RNA-seq data. In this paper, we investigate the performances of seven classifiers including support vector machine (SVM), random forest (RF), Naive Bayesian (NB), GBDT, logical regression (LR), decision tree (DT) and K-Nearest Neighbor (KNN) for solving the binary classification problems of GRN inference with single-cell RNA-seq data (Single_cell_GRN). In SVM, three different kernel functions (linear, polynomial and radial basis function) are utilized, respectively. Three real single-cell RNA-seq datasets from mouse and human are utilized. The experiment results prove that in most cases supervised learning methods (SVM, RF, NB, GBDT, LR, DT and KNN) perform better than unsupervised learning method (GENIE3) in terms of AUC. SVM, RF and KNN have the better performances than other four classifiers. In SVM, linear and polynomial kernels are more fit to model single-cell RNA-seq data.

The availability of the data has increased tremendously due to the excess usage of social media platforms like Twitter and Facebook. Due to the abundant availability of data, scientists, businesses, educationalists and other people working under different roles have started using Sentiment Analysis (SA) to get in-depth knowledge about the sentiments of the people regarding any topic of interest. There are many techniques to implement SA, and one of them is Machine Learning (ML). This study is focused on the comparison of ancient ML methods such as Naïve Bayes (NB), Decision Tree (DT), Support Vector Machine (SVM), and a modern method, i.e., Deep Learning (DL). The ML techniques are applied to a single dataset to compare their performance in terms of accuracy to understand how they perform against each other. The study found that DL performed the best with 96.41% accuracy followed by NB and SVM with 87.18% and 82.05% respectively. DT performed the poorest with 68.21% accuracy.

This article aims at performing maritime target classification in SAR images using machine learning (ML) and deep learning (DL) techniques. In particular, the targets of interest are oil platforms and ships located in the Campos Basin, Brazil. Two convolutional neural networks (CNNs), VGG-16 and VGG-19, were used for attribute extraction. The logistic regression (LR), random forest (RF), support vector machine (SVM), k-nearest neighbours (kNN), decision tree (DT), naive Bayes (NB), neural networks (NET), and AdaBoost (ADBST) schemes were considered for classification. The target classification methods were evaluated using polarimetric images obtained from the C-band synthetic aperture radar (SAR) system Sentinel-1. Classifiers are assessed by the accuracy indicator. The LR, SVM, NET, and stacking results indicate better performance, with accuracy ranging from 84.1% to 85.5%. The Kruskal–Wallis test shows a significant difference with the tested classifier, indicating that some classifiers present different accuracy results. The optimizations provide results with more significant accuracy gains, making them competitive with those shown in the literature. There is no exact combination of methods for SAR image classification that will always guarantee the best accuracy. The optimizations performed in this article were for the specific data set of the Campos Basin, and results may change depending on the data set format and the number of images.