Deception Detection in Turkish Hotel Reviews: A Comparative Study of Machine Learning and Deep Learning Approaches

With the increase in the culture of social media and netizen, every day, millions of comments are posted on the uploaded posts. The use of abusive language in user comments has been increased rapidly. Abusive language in online comments initiates cyber-bullying that targets individuals (celebrity, politician, and product) and a group of people (specific country, age, and religion). It is important to detect and analyze abusive language from online comments automatically. There have been several attempts in the literature to detect abusive language for English. In this study, we perform abusive language detection from Urdu and Roman Urdu comments using five diverse ML models (NB, SVM, IBK, Logistic, and JRip) and four DL models (CNN, LSTM, BLSTM, and CLSTM). We apply these models on a large dataset with ten thousands of Roman Urdu comments and a small dataset with more than two thousand comments of Urdu. Natural language constructs, English-like nature of Roman Urdu script, and Nastaleeq style of Urdu make it more challenging to process and classify the comments of both scripts using deep learning and machine learning approaches. From experiments, we find that the convolutional neural network outperforms the other models and achieves 96.2% and 91.4% accuracy on Urdu and Roman Urdu. Our results also reveal that the one-layer architectures of deep learning models give better results than two-layer architectures. Further, we compare the performance of deep learning models with five conventional machine learning models and conclude that deep learning models perform significantly better than machine learning models.

Plant diseases identification is generally done by visual evaluation. The diagnosis quality depends strongly on professional knowledge. However, agricultural expertise is not easily learned. To overcome this issue, automatic analysis and classification of plant diseases through image processing and artificial intelligence is an encouraging solution and can reduce the lack of agricultural knowledge. This alliance between image processing and machine learning and/or deep learning approaches has produced very good results in medical imaging and has enabled the development of robust systems to assist in the diagnosis of several diseases. Deep learning (DL) and Machine learning approaches became the most promising tools for image analysis and classification. The objective of our work is to conduct a comparative study between different deep learning architectures for the analysis and classification of plant diseases. Four DL-based architectures have been implemented such as MobileNet, AlexNet, Inception V3, and VGG16. Simulation results have shown that the MobileNet architecture, even simple, has allowed better classification accuracy.KeywordsPlant diseases detectionPlant diseases classificationImage classificationDeep learningMachine learningConvolutional neuron network

Deep learning, a state-of-the-art machine learning approach, has shown outstanding performance over traditional machine learning in identifying intricate structures in complex high-dimensional data, especially in the domain of computer vision. The application of deep learning to early detection and automated classification of Alzheimer's disease (AD) has recently gained considerable attention, as rapid progress in neuroimaging techniques has generated large-scale multimodal neuroimaging data. A systematic review of publications using deep learning approaches and neuroimaging data for diagnostic classification of AD was performed. A PubMed and Google Scholar search was used to identify deep learning papers on AD published between January 2013 and July 2018. These papers were reviewed, evaluated, and classified by algorithm and neuroimaging type, and the findings were summarized. Of 16 studies meeting full inclusion criteria, 4 used a combination of deep learning and traditional machine learning approaches, and 12 used only deep learning approaches. The combination of traditional machine learning for classification and stacked auto-encoder (SAE) for feature selection produced accuracies of up to 98.8% for AD classification and 83.7% for prediction of conversion from mild cognitive impairment (MCI), a prodromal stage of AD, to AD. Deep learning approaches, such as convolutional neural network (CNN) or recurrent neural network (RNN), that use neuroimaging data without pre-processing for feature selection have yielded accuracies of up to 96.0% for AD classification and 84.2% for MCI conversion prediction. The best classification performance was obtained when multimodal neuroimaging and fluid biomarkers were combined. Deep learning approaches continue to improve in performance and appear to hold promise for diagnostic classification of AD using multimodal neuroimaging data. AD research that uses deep learning is still evolving, improving performance by incorporating additional hybrid data types, such as—omics data, increasing transparency with explainable approaches that add knowledge of specific disease-related features and mechanisms.

We describe the sentiment analysis experiments that were performed on the Lithuanian Internet comment dataset using traditional machine learning (Naïve Bayes Multinomial—NBM and Support Vector Machine—SVM) and deep learning (Long Short-Term Memory—LSTM and Convolutional Neural Network—CNN) approaches. The traditional machine learning techniques were used with the features based on the lexical, morphological, and character information. The deep learning approaches were applied on the top of two types of word embeddings (Vord2Vec continuous bag-of-words with negative sampling and FastText). Both traditional and deep learning approaches had to solve the positive/negative/neutral sentiment classification task on the balanced and full dataset versions. The best deep learning results (reaching 0.706 of accuracy) were achieved on the full dataset with CNN applied on top of the FastText embeddings, replaced emoticons, and eliminated diacritics. The traditional machine learning approaches demonstrated the best performance (0.735 of accuracy) on the full dataset with the NBM method, replaced emoticons, restored diacritics, and lemma unigrams as features. Although traditional machine learning approaches were superior when compared to the deep learning methods; deep learning demonstrated good results when applied on the small datasets.

Since the COVID-19 outbreak, considering the people’s opinion has been perceived as the most crucial challenge for the government to combat the pandemic, such as implementing a national lockdown, instituting a quarantine procedure, providing health services, and more. Furthermore, the government made many critical decisions based on public opinion to combat coronavirus. Opinion mining or sentiment analysis has arisen as a method for mining people’s views on several issues using machine learning techniques. With the support of machine learning methods, this paper extracted the Indian people’s opinions on vaccines through Twitter tweets. More than four lakh vaccine-related tweets from May 04 to May 11, 2021, and from Aug 13 to Aug 21, 2021, were analyzed using state-of-the-art machine learning and deep learning approaches. The BERT and RoBERTa models produced promising results compared to other models on the collected twitter dataset.

Osteoporosis, arthritis, and fractures are examples of orthopedic illnesses that not only significantly impair patients' quality of life but also complicate and raise the expense of therapy. It has been discovered in recent years that the pathophysiology of orthopedic disorders is significantly influenced by the microbiota. By employing machine learning and deep learning techniques to conduct a thorough analysis of the disease-causing microbiome, we can enhance our comprehension of the pathophysiology of many illnesses and expedite the creation of novel treatment approaches. Today's science is undergoing a revolution because to the introduction of machine learning and deep learning technologies, and the field of biomedical research is no exception. The genesis, course, and management of orthopedic disorders are significantly influenced by pathogenic microbes. Orthopedic infection diagnosis and treatment are made more difficult by the lengthy and imprecise nature of traditional microbial detection and characterization techniques. These cutting-edge analytical techniques are offering previously unheard-of insights into the intricate relationships between orthopedic health and pathogenic microbes, opening up previously unimaginable possibilities for illness diagnosis, treatment, and prevention. The goal of biomedical research has always been to improve diagnostic and treatment methods while also gaining a deeper knowledge of the processes behind the onset and development of disease. Although traditional biomedical research methodologies have demonstrated certain limits throughout time, they nevertheless rely heavily on experimental data and expertise. This is the area in which deep learning and machine learning approaches excel. The advancements in machine learning (ML) and deep learning (DL) methodologies have enabled us to examine vast quantities of data and unveil intricate connections between microorganisms and orthopedic disorders. The importance of ML and DL in detecting, categorizing, and forecasting harmful microorganisms in orthopedic infectious illnesses is reviewed in this work.

Sales forecasting plays an important role in the modern financial system. It is used in the private and government financial institutions, companies, industries, factories, trading, etc. Due to the necessity of sales forecasting, this study focused the machine learning and deep learning approaches used for predicting the future sales and demands. These approaches accept the input as historical sales data and generate the response as future demands. From this survey it is observed that deep learning approaches are performed better than machine learning approaches in terms of prediction accuracy. In deep learning approaches, Convolutional Neural Network (CNN) can attain high prediction accuracy

Parkinson's disease (PD) is a chronic neurological condition that is growing in prevalence and manifests both motor and non-motor symptoms. Most PD patients have trouble speaking, writing, and walking during the early stages of the disease. Analysis of speech problems has been effective in identifying Parkinson's disease. According to studies, 90% of Parkinson's disease patients experience speech problems. Even though there is no known cure for Parkinson's disease, using the right medication at an early stage can greatly reduce the symptoms. One of the key categorization issues for the diagnosis of Parkinson's disease is the correct interpretation of speech signals. The major goal of this project is to use deep learning and machine learning approaches to predict and categorize PD patients at an early stage. A trustworthy dataset from the UCI repository for Parkinson disease has been used to evaluate the method's performance. Several classification models are successfully used in this study for classification tasks, including Extreme Gradient Boosting (XGBoost), Ada Boost, Light Gradient Boosting Machine, CatBoost, Gradient Boosting, Random Forest, Ridge, Decision Tree, Logistic Regression, K Neighbors, SVM - Linear Kernel, Naive Bayes, and deep neural networks (DNN1, DNN2, DNN3). The Extreme Gradient Boosting classifier achieved the greatest classification accuracy of 92.18% (among the machine learning classifiers). By using the chosen features as input, the three layer deep neural network (DNN2) has the best accuracy of 95.41% amongst deep learning techniques. The collected results indicate that deep neural networks performed better than machine learning methods.

Landslide is a devastating natural disaster with the frequent occurrence and tremendous destructive power. Once it happened, human society, the safety of life and property, and the natural environment would suffer enormous losses. The purpose of landslide research is to reduce landslide occurrence probability through manual intervention to some extent, in which landslide detection is one of the fundamental researches in this field. For state-of-art studies, the hotspot of landslide detection primarily focuses on Deep Learning (DL) and Machine Learning (ML) approaches. In this paper, we summarize the primary works in the field of landslide research firstly. Then the acquisition and usage of landslide data for DL and ML approaches are introduced. Next, the most frequently used evaluation indexes of object detection and image segmentation. Finally, the relevant progress of DL and ML approaches in landslide detection research are reviewed. Meanwhile, the challenges and future research directions in this field are further discussed.

Given the vast range of factors, including shape, size, skew, and orientation of handwritten numerals, their machine-based recognition is a difficult challenge for researchers in the pattern recognition field. Due to the abundance of curves and resembling shapes of the symbols, the recognition of Devnagari numerals can leverage the difficulty level of the recognition. The suggested low-classification-cost method for obtaining fine features from given numeral images used benchmark deep learning models, VGG-16Net, VGG-19Net, ResNet-50, and Inception-v3, to address these issues. Principal component analysis, a powerful dimensionality reduction method, was used to efficiently reduce the number of dimensions in the information that pre-trained deep convolutional neural network models provided. The method for improving recognition accuracy by fusing features was provided in the scheme. A machine learning algorithm: support vector machine was employed for the recognition task due to its capacity to distinguish between patterns belonging to distinct classes. The system was able to obtain a recognition accuracy of 99.72% and was effective in demonstrating the importance of ensemble machine learning and deep learning approaches.

One of the most important structures in the human body is the heart. It is at the focal point of the circulatory framework. Heart disease is a potentially fatal condition that can result in death or severe long-term impairment. Therefore, efficient techniques for uncovering hidden linkages and there are no clear patterns in e-health data. Medical diagnosis is a difficult task that is crucial to preserving lives, so it must be done correctly and quickly. To lower the cost of performing clinical tests, a suitable and precise computer-based automated decision support system is necessary. The use of machine learning in health analytics has been presented as a way to anticipate reliable patient data analysis. The information generated by the healthcare industry is not mined. In the medical industry, data mining techniques can be utilized to create an intelligent model employing data sets that include patient risk factors. The emergence of ideas and techniques for making use of data has surprised knowledge discovery in databases (KDD). This study delves into the usage of deep learning and machine learning approaches in disease diagnosis. In recent years, many data mining classifiers have been created to aid in the accurate and timely identification of illnesses. This study offers a heart attack forecasting model based on deep learning methodologies, notably Multi-Layer Perceptron (MLP), to anticipate a patient's likelihood of acquiring heart disease. MLP is an advanced type approach that employs the Artificial Neural Network's Deep Learning approach. Deep learning and data mining are used in the suggested approach to get reliable and error-free outcomes.

There are minimal published data on the relationship between personality traits and learning approaches among medical students. This study explored the causal-effect relationship of personality traits and learning approaches among medical students. A cross-sectional study was conducted on medical students and they responded to the Learning Approach Inventory and USM Personality Inventory to measure personality traits and learning approaches, respectively. A structural equation modelling was performed by AMOS 24 to test the causal-effect relationship of personality traits and learning approaches. Conscientiousness had a positive direct effect on deep learning approach, while neuroticism had negative direct effect on deep and strategic learning approaches. Extraversion, openness, and agreeableness had no significant link or effect on any learning approaches. Strategic learning approach had positive direct effect on deep learning approach and a mediator for surface learners on deep learning approach. Surface learning approach had a negative direct effect on deep learning approach. There was a significant relationship of specific personality traits and learning approaches. Conscientiousness and neuroticism had significant relationships with deep and strategic learning approaches. These findings enables medical educators to have a better understanding of the influence of personality traits on medical students’ learning approaches to learning tasks and their implications on instructional strategies.

Offensive language identification (OLI) in user generated text is automatic detection of any profanity, insult, obscenity, racism or vulgarity that degrades an individual or a group. It is helpful for hate speech detection, flame detection and cyber bullying. Due to immense growth of accessibility to social media, OLI helps to avoid abuse and hurts. In this paper, we present deep and traditional machine learning approaches for OLI. In deep learning approach, we have used bi-directional LSTM with different attention mechanisms to build the models and in traditional machine learning, TF-IDF weighting schemes with classifiers namely Multinomial Naive Bayes and Support Vector Machines with Stochastic Gradient Descent optimizer are used for model building. The approaches are evaluated on the OffensEval@SemEval2019 dataset and our team SSN_NLP submitted runs for three tasks of OffensEval shared task. The best runs of SSN_NLP obtained the F1 scores as 0.53, 0.48, 0.3 and the accuracies as 0.63, 0.84 and 0.42 for the tasks A, B and C respectively. Our approaches improved the base line F1 scores by 12%, 26% and 14% for Task A, B and C respectively.

Traders and investors are interested in accurately predicting cryptocurrency prices to increase returns and minimize risk. However, due to their uncertainty, volatility, and dynamism, forecasting crypto prices is a challenging time series analysis task. Researchers have proposed predictors based on statistical, machine learning (ML), and deep learning (DL) approaches, but the literature is limited. Indeed, it is narrow because it focuses on predicting only the prices of the few most famous cryptos. In addition, it is scattered because it compares different models on different cryptos inconsistently, and it lacks generality because solutions are overly complex and hard to reproduce in practice. The main goal of this paper is to provide a comparison framework that overcomes these limitations. We use this framework to run extensive experiments where we compare the performances of widely used statistical, ML, and DL approaches in the literature for predicting the price of five popular cryptocurrencies, i.e., XRP, Bitcoin (BTC), Litecoin (LTC), Ethereum (ETH), and Monero (XMR). To the best of our knowledge, we are also the first to propose using the temporal fusion transformer (TFT) on this task. Moreover, we extend our investigation to hybrid models and ensembles to assess whether combining single models boosts prediction accuracy. Our evaluation shows that DL approaches are the best predictors, particularly the LSTM, and this is consistently true across all the cryptos examined. LSTM reaches an average RMSE of 0.0222 and MAE of 0.0173, respectively, 2.7% and 1.7% better than the second-best model. To ensure reproducibility and stimulate future research contribution, we share the dataset and the code of the experiments.

BackgroundBrain tumor classification using Magnetic Resonance Imaging (MRI) is crucial for diagnosis and treatment planning. The differentiation between malignant and benign brain tumors and their subtypes remains a challenging task that can benefit from advanced computational techniques.PurposeThis study uses an MRI dataset to explore the effectiveness of deep learning (DL) and machine learning (ML) approaches for classifying brain tumors.Materials and methodsA dataset comprising 1200 DICOM brain tumor MRI images, representing malignant and benign tumors with six subtypes, was prepared. Each image was converted to a 512 × 512-pixel digital format, selecting 200 images per tumor class. Image quality was enhanced using sharpening algorithms and mean filtering. The proposed edge refined binary histogram segmentation (ER-BHS) was applied to extract hybrid features from the regions of interest. Feature optimization through a correlation-based method reduced the dataset to 11 key features. Multiple classifiers, including DL, neural networks, and ML models, were evaluated on the optimized dataset using 10-fold cross-validation.ResultsAmong the tested models, the random committee (RC) classifier demonstrated superior performance, achieving an accuracy of 98.61% on the optimized hybrid brain tumor MRI dataset. Overall, DL and ML methods effectively automated brain tumor classification.ConclusionThe promising results affirm the potential of DL and ML approaches to enhance medical image analysis and improve diagnostic accuracy in brain tumor classification, potentially revolutionizing clinical workflows.