scholarly journals Benchmarking Deep Learning Methods for Aspect Level Sentiment Classification

2021 ◽  
Vol 11 (22) ◽  
pp. 10542
Author(s):  
Tanu Sharma ◽  
Kamaldeep Kaur
Keyword(s):  
Deep Learning ◽  
Performance Metrics ◽  
Statistical Tests ◽  
Sentiment Classification ◽  
Ranking Methods ◽  
Learning Methods ◽  
Statistical Comparison ◽  
Benchmark Datasets ◽  
Overall Performance ◽  
Selection Of

With the advancements in processing units and easy availability of cloud-based GPU servers, many deep learning-based methods have been proposed for Aspect Level Sentiment Classification (ALSC) literature. With this increase in the number of deep learning methods proposed in ALSC literature, it has become difficult to ascertain the performance difference of one method over the other. To this end, our study provides a statistical comparison of the performance of 35 recent deep learning methods with respect to three performance metrics-Accuracy, Macro F1 score, and Time. The methods are evaluated for eight benchmark datasets. In this study, the statistical comparison is based on Friedman, Nemenyi, and Wilcoxon tests. As per the results of statistical tests, the top-ranking methods could not significantly outperform several other methods in terms of Accuracy and Macro F1 score and performed poorly on-time metric. However, the time taken by any method is crucial to analyze the overall performance. Thus, this study aids the selection of the Deep Learning method, which maximizes the accuracy and Macro F1 score and takes minimal time. Our study also establishes a framework for validating the performance of new and alternate methods in ALSC that can be helpful for researchers and practitioners working in this area.

scholarly journals Selection of acoustic modeling unit for Tibetan speech recognition based on deep learning

2021 ◽  
Vol 336 ◽  
pp. 06014
Author(s):  
Baojia Gong ◽  
Rangzhuoma Cai ◽  
Zhijie Cai ◽  
Yuntao Ding ◽  
Maozhaxi Peng
Keyword(s):  
Deep Learning ◽  
Speech Recognition ◽  
Experimental Verification ◽  
Character Segmentation ◽  
Acoustic Modeling ◽  
Overall Performance ◽  
Primary Problem ◽  
Selection Of

The selection of the speech recognition modeling unit is the primary problem of acoustic modeling in speech recognition, and different acoustic modeling units will directly affect the overall performance of speech recognition. This paper designs the Tibetan character segmentation and labeling model and algorithm flow for the purpose of solving the problem of selecting the acoustic modeling unit in Tibetan speech recognition by studying and analyzing the deficiencies of the existing acoustic modeling units in Tibetan speech recognition. After experimental verification, the Tibetan character segmentation and labeling model and algorithm achieved good performance of character segmentation and labeling, and the accuracy of Tibetan character segmentation and labeling reached 99.98%, respectively.

scholarly journals Embedding knowledge on ontology into the corpus by topic to improve the performance of deep learning methods in sentiment analysis

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Duy Ngoc Nguyen ◽  
Tuoi Thi Phan ◽  
Phuc Do
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Sentiment Classification ◽  
Test Results ◽  
High Confidence ◽  
Knowledge Processing ◽  
Learning Methods ◽  
Novel Approach

AbstractSentiment classification, which uses deep learning algorithms, has achieved good results when tested with popular datasets. However, it will be challenging to build a corpus on new topics to train machine learning algorithms in sentiment classification with high confidence. This study proposes a method that processes embedding knowledge in the ontology of opinion datasets called knowledge processing and representation based on ontology (KPRO) to represent the significant features of the dataset into the word embedding layer of deep learning algorithms in sentiment classification. Unlike the methods that lexical encode or add information to the corpus, this method adds presentation of raw data based on the expert’s knowledge in the ontology. Once the data has a rich knowledge of the topic, the efficiency of the machine learning algorithms is significantly enhanced. Thus, this method is appliable to embed knowledge in datasets in other languages. The test results show that deep learning methods achieved considerably higher accuracy when trained with the KPRO method’s dataset than when trained with datasets not processed by this method. Therefore, this method is a novel approach to improve the accuracy of deep learning algorithms and increase the reliability of new datasets, thus making them ready for mining.

scholarly journals Combining Post Sentiments and User Participation for Extracting Public Stances from Twitter

2020 ◽  
Vol 10 (22) ◽  
pp. 8035
Author(s):  
Jenq-Haur Wang ◽  
Ting-Wei Liu ◽  
Xiong Luo
Keyword(s):  
Social Media ◽  
Deep Learning ◽  
Short Term Memory ◽  
Absolute Error ◽  
User Participation ◽  
Sentiment Classification ◽  
Context Dependency ◽  
Long Distance ◽  
Learning Methods ◽  
The Public

With the wide popularity of social media, it’s becoming more convenient for people to express their opinions online. To better understand what the public think about a topic, sentiment classification techniques have been widely used to estimate the overall orientation of opinions in post contents. However, users might have various degrees of influence depending on their participation in discussions on different topics. In this paper, we address the issues of combining sentiment classification and link analysis techniques for extracting stances of the public from social media. Since social media posts are usually very short, word embedding models are first used to learn different word usages in various contexts. Then, deep learning methods such as Long Short-Term Memory (LSTM) are used to learn the long-distance context dependency among words for better estimation of sentiments. Third, we consider the major user participation in popular social media by adjusting the users weights to reflect their relative influence in user-post interaction graphs. Finally, we combine post sentiments and user influences into a total opinion score for extracting public stances. In the experiments, we evaluated the performance of our proposed approach for tweets about the 2016 U.S. Presidential Election. The best performance of sentiment classification can be observed with an F-measure of 72.97% for LSTM classifiers. This shows the effectiveness of deep learning methods in learning word usage in social media contexts. The experimental results on stance extraction showed the best performance of 0.68% Mean Absolute Error (MAE) in aggregating public stances on election candidates. This shows the potential of combining tweet sentiments and user participation structures for extracting the aggregate stances of the public on popular topics. Further investigation is needed to verify the performance in different social media sources.

scholarly journals Using human pose information for handgun detection

2021 ◽  
Author(s):  
Alberto Velasco-Mata ◽  
Jesus Ruiz-Santaquiteria ◽  
Noelia Vallez ◽  
Oscar Deniz
Keyword(s):  
Deep Learning ◽  
Public Spaces ◽  
Learning Methods ◽  
Maximum Improvement ◽  
Overall Performance ◽  
Human Pose ◽  
Grayscale Images

AbstractFast automatic handgun detection can be very useful to avoid or mitigate risks in public spaces. Detectors based on deep learning methods have been proposed in the literature to trigger an alarm if a handgun is detected in the image. However, those detectors are solely based on the weapon appearance on the image. In this work, we propose to combine the detector with the individual’s pose information in order to improve overall performance. To this end, a model that integrates grayscale images from the output of the handgun detector and heatmap-like images that represent pose is proposed. The results show an improvement over the original handgun detector. The proposed network provides a maximum improvement of a 17.5% in AP of the proposed combinational model over the baseline handgun detector.

scholarly journals Performance of a Novel Chaotic Firefly Algorithm with Enhanced Exploration for Tackling Global Optimization Problems: Application for Dropout Regularization

Mathematics ◽  
2021 ◽  
Vol 9 (21) ◽  
pp. 2705
Author(s):  
Nebojsa Bacanin ◽  
Ruxandra Stoean ◽  
Miodrag Zivkovic ◽  
Aleksandar Petrovic ◽  
Tarik A. Rashid ◽  
...  
Keyword(s):  
Neural Networks ◽  
Global Optimization ◽  
Deep Learning ◽  
Firefly Algorithm ◽  
Optimization Problems ◽  
Wide Spectrum ◽  
Statistical Tests ◽  
Dropout Rate ◽  
Original Method ◽  
Benchmark Datasets

Swarm intelligence techniques have been created to respond to theoretical and practical global optimization problems. This paper puts forward an enhanced version of the firefly algorithm that corrects the acknowledged drawbacks of the original method, by an explicit exploration mechanism and a chaotic local search strategy. The resulting augmented approach was theoretically tested on two sets of bound-constrained benchmark functions from the CEC suites and practically validated for automatically selecting the optimal dropout rate for the regularization of deep neural networks. Despite their successful applications in a wide spectrum of different fields, one important problem that deep learning algorithms face is overfitting. The traditional way of preventing overfitting is to apply regularization; the first option in this sense is the choice of an adequate value for the dropout parameter. In order to demonstrate its ability in finding an optimal dropout rate, the boosted version of the firefly algorithm has been validated for the deep learning subfield of convolutional neural networks, with respect to five standard benchmark datasets for image processing: MNIST, Fashion-MNIST, Semeion, USPS and CIFAR-10. The performance of the proposed approach in both types of experiments was compared with other recent state-of-the-art methods. To prove that there are significant improvements in results, statistical tests were conducted. Based on the experimental data, it can be concluded that the proposed algorithm clearly outperforms other approaches.

scholarly journals Deep Learning for Social Media Sentiment Analysis

MATEMATIKA ◽  
2020 ◽  
Vol 36 (2) ◽  
pp. 99-111
Author(s):  
Kartika Fithriasari ◽  
Saidah Zahrotul Jannah ◽  
Zakya Reyhana
Keyword(s):  
Neural Network ◽  
Social Media ◽  
Deep Learning ◽  
Sentiment Analysis ◽  
Characteristic Curve ◽  
Sentiment Classification ◽  
Radio Station ◽  
Learning Methods ◽  
Text Data ◽  
Media Sentiment

Social media is used as a tool by many people to express their opinions. Sentiment analysis for social media is very important, as it allows information to be obtained about public opinion on government performance. The goal of this research is to learn about the opinions of Surabaya citizens, using deep learning methods. The data are extracted from the official Twitter accounts of the Surabaya government and a private radio station in Surabaya. The data are grouped into two categories: positive and negative sentiments. This research is conducted in three steps: data pre-processing, sentiment classification, and visualization. Data pre-processing is required before modelling approaches are applied. It is used to transform the unstructured text data into structured data. The data pre-processing consists of case folding, tokenizing, and the removal of stop words. Deep learning methods are then applied to the data. A Backpropagation Neural Network (BNN) and a Convolutional Neural Network (CNN) are used to perform the sentiment classification. The BNN and CNN are compared using various metrics, such as precision, sensitivity, and area under the receiver operating characteristic curve (AUC). A word cloud is then used to visualize the data and find the most frequent words in each class. The results show that the sentiment classification with CNN is better than that with the BNN because the values for the precision, sensitivity and AUC are higher.

scholarly journals End-to-End Adversarial Memory Network for Cross-domain Sentiment Classification

2017 ◽  
Author(s):  
Zheng Li ◽  
Yu Zhang ◽  
Ying Wei ◽  
Yuxiang Wu ◽  
Qiang Yang
Keyword(s):  
Deep Learning ◽  
Domain Adaptation ◽  
Sentiment Classification ◽  
Classification Error ◽  
Target Domain ◽  
Learning Methods ◽  
Cross Domain ◽  
Memory Network ◽  
End To End ◽  
Two Parameter

Domain adaptation tasks such as cross-domain sentiment classification have raised much attention in recent years. Due to the domain discrepancy, a sentiment classifier trained in a source domain may not work well when directly applied to a target domain. Traditional methods need to manually select pivots, which behave in the same way for discriminative learning in both domains. Recently, deep learning methods have been proposed to learn a representation shared by domains. However, they lack the interpretability to directly identify the pivots. To address the problem, we introduce an end-to-end Adversarial Memory Network (AMN) for cross-domain sentiment classification. Unlike existing methods, our approach can automatically capture the pivots using an attention mechanism. Our framework consists of two parameter-shared memory networks: one is for sentiment classification and the other is for domain classification. The two networks are jointly trained so that the selected features minimize the sentiment classification error and at the same time make the domain classifier indiscriminative between the representations from the source or target domains. Moreover, unlike deep learning methods that cannot tell us which words are the pivots, our approach can offer a direct visualization of them. Experiments on the Amazon review dataset demonstrate that our approach can significantly outperform state-of-the-art methods.

scholarly journals DLAB - Deep learning methods for structure-based virtual screening of antibodies

2021 ◽  
Author(s):  
Constantin Schneider ◽  
Andrew Buchanan ◽  
Bruck Taddese ◽  
Charlotte M. Deane
Keyword(s):  
Deep Learning ◽  
Virtual Screening ◽  
Drug Discovery Process ◽  
Learning Methods ◽  
Drug Candidates ◽  
Specific Antigens ◽  
High Throughput Screens ◽  
Selection Of

AbstractAntibodies are one of the most important classes of pharmaceuticals, with over 80 approved molecules currently in use against a wide variety of diseases. The drug discovery process for antibody therapeutic candidates however is time- and cost-intensive and heavily reliant on in-vivo and in-vitro high throughput screens. Here, we introduce a framework for structure-based deep learning for antibodies (DLAB) which can virtually screen putative binding antibodies against antigen targets of interest. DLAB is built to be able to predict antibody-antigen binding for antigens with no known antibody binders.We demonstrate that DLAB can be used both to improve antibody-antigen docking and structure-based virtual screening of antibody drug candidates. DLAB enables improved pose ranking for antibody docking experiments as well as selection of antibody-antigen pairings for which accurate poses are generated and correctly ranked. DLAB also outperforms baseline methods at identifying binding antibodies against specific antigens in a series of case studies. Our results demonstrate the promise of deep learning methods for structure-based virtual screening of antibodies.

scholarly journals Unsupervised phenotypic analysis of cellular images with multi-scale convolutional neural networks

2018 ◽  
Author(s):  
William J. Godinez ◽  
Imtiaz Hossain ◽  
Xian Zhang
Keyword(s):  
Neural Networks ◽  
Deep Learning ◽  
Large Scale ◽  
Cellular Imaging ◽  
Phenotypic Analysis ◽  
Learning Methods ◽  
Multi Scale ◽  
Full Resolution ◽  
Benchmark Datasets ◽  
Microscopy Images

AbstractLarge-scale cellular imaging and phenotyping is a widely adopted strategy for understanding biological systems and chemical perturbations. Quantitative analysis of cellular images for identifying phenotypic changes is a key challenge within this strategy, and has recently seen promising progress with approaches based on deep neural networks. However, studies so far require either pre-segmented images as input or manual phenotype annotations for training, or both. To address these limitations, we have developed an unsupervised approach that exploits the inherent groupings within cellular imaging datasets to define surrogate classes that are used to train a multi-scale convolutional neural network. The trained network takes as input full-resolution microscopy images, and, without the need for segmentation, yields as output feature vectors that support phenotypic profiling. Benchmarked on two diverse benchmark datasets, the proposed approach yields accurate phenotypic predictions as well as compound potency estimates comparable to the state-of-the-art. More importantly, we show that the approach identifies novel cellular phenotypes not included in the manual annotation nor detected by previous studies.Author summaryCellular microscopy images provide detailed information about how cells respond to genetic or chemical treatments, and have been widely and successfully used in basic research and drug discovery. The recent breakthrough of deep learning methods for natural imaging recognition tasks has triggered the development and application of deep learning methods to cellular images to understand how cells change upon perturbation. Although successful, deep learning studies so far either can only take images of individual cells as input or require human experts to label a large amount of images. In this paper, we present an unsupervised deep learning approach that, without any human annotation, analyzes directly full-resolution microscopy images displaying typically hundreds of cells. We apply the approach to two benchmark datasets, and show that the approach identifies novel visual phenotypes not detected by previous studies.

scholarly journals Modeling Traders’ Behavior with Deep Learning and Machine Learning Methods: Evidence from BIST 100 Index

Complexity ◽  
2020 ◽  
Vol 2020 ◽  
pp. 1-16
Author(s):  
Afan Hasan ◽  
Oya Kalıpsız ◽  
Selim Akyokuş
Keyword(s):  
Machine Learning ◽  
Deep Learning ◽  
Financial Market ◽  
Performance Metrics ◽  
Confusion Matrix ◽  
Support Vector ◽  
Human Beings ◽  
Learning Methods ◽  
Technical Indicators ◽  
Machine Learning Methods

Although the vast majority of fundamental analysts believe that technical analysts’ estimates and technical indicators used in these analyses are unresponsive, recent research has revealed that both professionals and individual traders are using technical indicators. A correct estimate of the direction of the financial market is a very challenging activity, primarily due to the nonlinear nature of the financial time series. Deep learning and machine learning methods on the other hand have achieved very successful results in many different areas where human beings are challenged. In this study, technical indicators were integrated into the methods of deep learning and machine learning, and the behavior of the traders was modeled in order to increase the accuracy of forecasting of the financial market direction. A set of technical indicators has been examined based on their application in technical analysis as input features to predict the oncoming (one-period-ahead) direction of Istanbul Stock Exchange (BIST100) national index. To predict the direction of the index, Deep Neural Network (DNN), Support Vector Machine (SVM), Random Forest (RF), and Logistic Regression (LR) classification techniques are used. The performance of these models is evaluated on the basis of various performance metrics such as confusion matrix, compound return, and max drawdown.

Sign in / Sign up

Export Citation Format

Share Document