Search results for: STS benchmark dataset

Authors: Jawharah Alasmari

Abstract:

In verb inflection, tense marks past/present/future action, and aspect marks progressive/continues perfect/completed actions. This usage and meaning of tense and aspect differ in Arabic and English. In this research, we applied data mining methods to test the predictive function of candidate features by using our dataset of Arabic verbs in-context, and their 7 translations. Weka machine learning classifiers is used in this experiment in order to examine the key features that can be used to provide guidance to enable a translator’s appropriate English translation of the Arabic verb tense and aspect.

Keywords: Arabic verb, English translations, mining methods, Weka software

Procedia PDF Downloads 258

Authors: Xinyi Xu

Abstract:

This study conducts a detailed examination of Asian stock markets over the period from 2008 to 2023, with a focus on the dynamics of cointegration and their relevance for long-term investment strategies. Specifically, we assess the co-movement and potential for pairs trading—a strategy where investors take opposing positions on two stocks, indices, or financial instruments that historically move together. For example, we explore the relationship between the Nikkei 225 (N225), Japan’s benchmark stock index, and the Straits Times Index (STI) of Singapore, as well as the relationship between the Korea Composite Stock Price Index (KS11) and the STI. The methodology includes tests for normality, stationarity, cointegration, and the application of Vector Error Correction Modeling (VECM). Our findings reveal significant long-term relationships between these pairs, indicating opportunities for pairs trading strategies. Furthermore, the research underscores the challenges posed by model instability and the influence of major global incidents, which are identified as structural breaks. These findings pave the way for further exploration into the intricacies of financial market dynamics.

Keywords: normality tests, stationarity, cointegration, VECM, pairs trading

Procedia PDF Downloads 37

Authors: Ghazwan Al-Haji

Abstract:

Traffic safety has become a major concern in Southeast Asia due to the increasing number of road accidents resulting in fatalities and injuries. Southeast Asia has one of the highest road traffic fatality rates in the world, in terms of both population and number of cars, nearly six times higher than the EU region. One of the reasons for this concerning trend is the increasing share of motorcycles as a form of transportation throughout Southeast Asia. The purpose of this study is to benchmark traffic safety situations and statistics in six countries in Asia and the EU, which Indonesia, Malaysia, Vietnam, Italy, Portugal and Sweden. The research will assess the priorities and causes of road accidents in the target nations. Further, the study will analyze the existing practices and promote best practices that can be implemented toward safer roads in Asian target countries. In order to achieve this goal, the study categorizes various factors contributing to traffic accidents and best practices into 4 pillars (Safer Behavior, Safer Roads, Safer Vehicles and Road Safety Management). The result of the study consists of a list of recommendations that can be applied by policymakers to promote safer roads in Asia towards 2030. The study is co-financed by the EU project ASIASAFE.

Keywords: traffic safety, ASIASAFE, Southeast Asia, EU project

Procedia PDF Downloads 53

Authors: Dangut Maren David, Skaf Zakwan

Abstract:

Adequate monitoring of vehicle component in other to obtain high uptime is the goal of predictive maintenance, the major challenge faced by businesses in industries is the significant cost associated with a delay in service delivery due to system downtime. Most of those businesses are interested in predicting those problems and proactively prevent them in advance before it occurs, which is the core advantage of Prognostic Health Management (PHM) application. The recent emergence of industry 4.0 or industrial internet of things (IIoT) has led to the need for monitoring systems activities and enhancing system-to-system or component-to- component interactions, this has resulted to a large generation of data known as big data. Analysis of big data represents an increasingly important, however, due to complexity inherently in the dataset such as imbalance classification problems, it becomes extremely difficult to build a model with accurate high precision. Data-driven predictive modeling for condition-based maintenance (CBM) has recently drowned research interest with growing attention to both academics and industries. The large data generated from industrial process inherently comes with a different degree of complexity which posed a challenge for analytics. Thus, imbalance classification problem exists perversely in industrial datasets which can affect the performance of learning algorithms yielding to poor classifier accuracy in model development. Misclassification of faults can result in unplanned breakdown leading economic loss. In this paper, an advanced approach for handling imbalance classification problem is proposed and then a prognostic model for predicting aircraft component replacement is developed to predict component replacement in advanced by exploring aircraft historical data, the approached is based on hybrid ensemble-based method which improves the prediction of the minority class during learning, we also investigate the impact of our approach on multiclass imbalance problem. We validate the feasibility and effectiveness in terms of the performance of our approach using real-world aircraft operation and maintenance datasets, which spans over 7 years. Our approach shows better performance compared to other similar approaches. We also validate our approach strength for handling multiclass imbalanced dataset, our results also show good performance compared to other based classifiers.

Keywords: prognostics, data-driven, imbalance classification, deep learning

Procedia PDF Downloads 157

Authors: Ágoston Nagy

Abstract:

This paper introduces a method of using unsupervised machine learning to visualize the feature space of a dataset in 2D, in order to find most characteristic segments in the set. After dimension reduction, users can select clusters by manual drawing. Selected clusters are recorded into a data model that is used for later predictions, based on realtime data. Predictions are made with supervised learning, using Gesture Recognition Toolkit. The paper introduces two example applications: a semantic audio organizer for analyzing incoming sounds, and a gesture database organizer where gestural data (recorded by a Leap motion) is visualized for further manipulation.

Keywords: gesture recognition, machine learning, real-time interaction, visualization

Procedia PDF Downloads 334

Authors: Ghada Moussa, Mamoud Owais

Abstract:

Transit Route Network Design Problem (TrNDP) is the most important component in Transit planning, in which the overall cost of the public transportation system highly depends on it. The main purpose of this study is to develop a novel solution methodology for the TrNDP, which goes beyond pervious traditional sophisticated approaches. The novelty of the solution methodology, adopted in this paper, stands on the deterministic operators which are tackled to construct bus routes. The deterministic manner of the TrNDP solution relies on using linear and integer mathematical formulations that can be solved exactly with their standard solvers. The solution methodology has been tested through Mandl’s benchmark network problem. The test results showed that the methodology developed in this research is able to improve the given network solution in terms of number of constructed routes, direct transit service coverage, transfer directness and solution reliability. Although the set of routes resulted from the methodology would stand alone as a final efficient solution for TrNDP, it could be used as an initial solution for meta-heuristic procedures to approach global optimal. Based on the presented methodology, a more robust network optimization tool would be produced for public transportation planning purposes.

Keywords: integer programming, transit route design, transportation, urban planning

Procedia PDF Downloads 251

Authors: William Whiteley, Jens Gregor

Abstract:

In recent developments of direct neural network based positron emission tomography (PET) reconstruction, two prominent architectures have emerged for converting measurement data into images: 1) networks that contain fully-connected layers; and 2) networks that primarily use a convolutional encoder-decoder architecture. In this paper, we present a multi-modal direct PET reconstruction method called MDPET, which is a hybrid approach that combines the advantages of both types of networks. MDPET processes raw data in the form of sinograms and histo-images in concert with attenuation maps to produce high quality multi-slice PET images (e.g., 8x440x440). MDPET is trained on a large whole-body patient data set and evaluated both quantitatively and qualitatively against target images reconstructed with the standard PET reconstruction benchmark of iterative ordered subsets expectation maximization. The results show that MDPET outperforms the best previously published direct neural network methods in measures of bias, signal-to-noise ratio, mean absolute error, and structural similarity.

Keywords: deep learning, image reconstruction, machine learning, neural network, positron emission tomography

Procedia PDF Downloads 95

Authors: Shengnan Chen, Shuhua Wang

Abstract:

Successful production of hydrocarbon from unconventional tight oil reserves has changed the energy landscape in North America. The oil contained within these reservoirs typically will not flow to the wellbore at economic rates without assistance from advanced horizontal well and multi-stage hydraulic fracturing. Efficient and economic development of these reserves is a priority of society, government, and industry, especially under the current low oil prices. Meanwhile, society needs technological and process innovations to enhance oil recovery while concurrently reducing environmental impacts. Recently, big data analysis and artificial intelligence become very popular, developing data-driven insights for better designs and decisions in various engineering disciplines. However, the application of data mining in petroleum engineering is still in its infancy. The objective of this research aims to apply intelligent data analysis and data-driven models to exploit unconventional oil reserves both efficiently and economically. More specifically, a comprehensive database including the reservoir geological data, reservoir geophysical data, well completion data and production data for thousands of wells is firstly established to discover the valuable insights and knowledge related to tight oil reserves development. Several data analysis methods are introduced to analysis such a huge dataset. For example, K-means clustering is used to partition all observations into clusters; principle component analysis is applied to emphasize the variation and bring out strong patterns in the dataset, making the big data easy to explore and visualize; exploratory factor analysis (EFA) is used to identify the complex interrelationships between well completion data and well production data. Different data mining techniques, such as artificial neural network, fuzzy logic, and machine learning technique are then summarized, and appropriate ones are selected to analyze the database based on the prediction accuracy, model robustness, and reproducibility. Advanced knowledge and patterned are finally recognized and integrated into a modified self-adaptive differential evolution optimization workflow to enhance the oil recovery and maximize the net present value (NPV) of the unconventional oil resources. This research will advance the knowledge in the development of unconventional oil reserves and bridge the gap between the big data and performance optimizations in these formations. The newly developed data-driven optimization workflow is a powerful approach to guide field operation, which leads to better designs, higher oil recovery and economic return of future wells in the unconventional oil reserves.

Keywords: big data, artificial intelligence, enhance oil recovery, unconventional oil reserves

Procedia PDF Downloads 271

Authors: Christodoulou Christos, Politis Anastasios

Abstract:

A large majority of users favoured to wireless LAN connection since it was so simple to use. A wireless network can be the target of numerous attacks. Class hijacking is a well-known attack that is fairly simple to execute and has significant repercussions on users. The statistical flow analysis based on machine learning (ML) techniques is a promising categorization methodology. In a given dataset, which in the context of this paper is a collection of components representing frames belonging to various flows, machine learning (ML) can offer a technique for identifying and characterizing structural patterns. It is possible to classify individual packets using these patterns. It is possible to identify fraudulent conduct, such as class hijacking, and take necessary action as a result. In this study, we explore a way to use machine learning approaches to thwart this attack.

Keywords: wireless lan, quality of service, machine learning, class hijacking, EDCA remapping

Procedia PDF Downloads 43

Authors: Soheila Sadeghi

Abstract:

In the dynamic landscape of project management, scope changes are an inevitable reality that can significantly impact project performance. These changes, whether initiated by stakeholders, external factors, or internal project dynamics, can lead to cost overruns and schedule delays. Accurately predicting the consequences of these changes is crucial for effective project control and informed decision-making. This study aims to develop predictive models to estimate the impact of scope changes on project cost and schedule using machine learning techniques. The research utilizes a comprehensive dataset containing detailed information on project tasks, including the Work Breakdown Structure (WBS), task type, productivity rate, estimated cost, actual cost, duration, task dependencies, scope change magnitude, and scope change timing. Multiple machine learning models are developed and evaluated to predict the impact of scope changes on project cost and schedule. These models include Linear Regression, Decision Tree, Ridge Regression, Random Forest, Gradient Boosting, and XGBoost. The dataset is split into training and testing sets, and the models are trained using the preprocessed data. Cross-validation techniques are employed to assess the robustness and generalization ability of the models. The performance of the models is evaluated using metrics such as Mean Squared Error (MSE) and R-squared. Residual plots are generated to assess the goodness of fit and identify any patterns or outliers. Hyperparameter tuning is performed to optimize the XGBoost model and improve its predictive accuracy. The feature importance analysis reveals the relative significance of different project attributes in predicting the impact on cost and schedule. Key factors such as productivity rate, scope change magnitude, task dependencies, estimated cost, actual cost, duration, and specific WBS elements are identified as influential predictors. The study highlights the importance of considering both cost and schedule implications when managing scope changes. The developed predictive models provide project managers with a data-driven tool to proactively assess the potential impact of scope changes on project cost and schedule. By leveraging these insights, project managers can make informed decisions, optimize resource allocation, and develop effective mitigation strategies. The findings of this research contribute to improved project planning, risk management, and overall project success.

Keywords: cost impact, machine learning, predictive modeling, schedule impact, scope changes

Procedia PDF Downloads 18

Authors: Jeh Patel, Chandrahas Paidi, Ahmed Hambaba

Abstract:

Traditional methods for analyzing athlete movement often lack the detail and immediacy required for optimal training. This project aims to address this limitation by developing a Real-time human pose estimation system specifically designed to enhance athlete training across various sports. This system leverages the power of convolutional neural networks (CNNs) to provide a comprehensive and immediate analysis of an athlete’s movement patterns during training sessions. The core architecture utilizes dilated convolutions to capture crucial long-range dependencies within video frames. Combining this with the robust encoder-decoder architecture to further refine pose estimation accuracy. This capability is essential for precise joint localization across the diverse range of athletic poses encountered in different sports. Furthermore, by quantifying movement efficiency, power output, and range of motion, the system provides data-driven insights that can be used to optimize training programs. Pose estimation data analysis can also be used to develop personalized training plans that target specific weaknesses identified in an athlete’s movement patterns. To overcome the limitations posed by outdoor environments, the project employs strategies such as multi-camera configurations or depth sensing techniques. These approaches can enhance pose estimation accuracy in challenging lighting and occlusion scenarios, where pose estimation accuracy in challenging lighting and occlusion scenarios. A dataset is collected From the labs of Martin Luther King at San Jose State University. The system is evaluated through a series of tests that measure its efficiency and accuracy in real-world scenarios. Results indicate a high level of precision in recognizing different poses, substantiating the potential of this technology in practical applications. Challenges such as enhancing the system’s ability to operate in varied environmental conditions and further expanding the dataset for training were identified and discussed. Future work will refine the model’s adaptability and incorporate haptic feedback to enhance the interactivity and richness of the user experience. This project demonstrates the feasibility of an advanced pose detection model and lays the groundwork for future innovations in assistive enhancement technologies.

Keywords: computer vision, deep learning, human pose estimation, U-NET, CNN

Procedia PDF Downloads 18

Authors: Shahpar Yilmaz, Yasir Nadeem, Syed A. Mehdi

Abstract:

Searching for a record in a dataset is always a frequent task for any data structure-related application. Hence, a fast and efficient algorithm for the approach has its importance in yielding the quickest results and enhancing the overall productivity of the company. Interpolation search is one such technique used to search through a sorted set of elements. This paper proposes a new algorithm, an advancement over interpolation search for the application of search over a sorted array. Pattern Recognition Search or PR Search (PRS), like interpolation search, is a pattern-based divide and conquer algorithm whose objective is to reduce the sample size in order to quicken the process and it does so by treating the array as a perfect arithmetic progression series and thereby deducing the key element’s position. We look to highlight some of the key drawbacks of interpolation search, which are accounted for in the Pattern Recognition Search.

Keywords: array, complexity, index, sorting, space, time

Procedia PDF Downloads 220

Authors: Siva Sankara Arudra, Kamaruzaman Baharuddin, Ir. Ahmed Fadzil Mustafa Kamal, Ir. Abdul Latif Mohamed

Abstract:

The advancement of electrical and electronics technologies has rewarded the oil and gas industry with great opportunities to embed more environmentally solutions into design. Most offshore oil and gas producers have their engineering and production asset goals to promote greater use of environmentally friendly compression system technologies to eliminate hazardous emissions from conventional gas compressor drivers. Therefore, this paper comprehensively elaborates the parametric study conducted in integrating the latest electrical and electronics drives technology into the existing compression system. This study was conducted in aspects of layout, reliability & availability, maintainability, emission, and cost. An existing offshore facility that utilized gas turbines as the driver for gas compression was set as Conventional Case for this study. The Electrification Case will utilize electric motor drives as the driver for the compression system. Findings from this study indicate more advantages in driver electrification compared to conventional compression systems. The findings of this paper can be set as a benchmark for future offshore driver selection for gas compression systems of similar operating parameters and power range.

Keywords: turbomachinery, electrification, emission, compression system

Procedia PDF Downloads 131

Authors: Xu Zhang

Abstract:

This article unveils an indicator of language contact between English and Cantonese in one of the Outer Circle Englishes, Hong Kong (HK) English, through an empirical investigation into 1000 tokens from the Global Web-based English (GloWbE) corpus, employing frequency analysis and logistic regression analysis. It is perceived that Cantonese and general Chinese are contextually marked by an integral underlying thinking pattern. Chinese speakers exhibit a reliance on semantic context over syntactic rules and lexical forms. This linguistic trait carries over to their use of English, affording greater flexibility to formal elements in constructing English sentences. The study focuses on the syntactic positioning of the focusing subjunct ‘also’, a linguistic element used to add new or contrasting prominence to specific sentence constituents. The English language generally allows flexibility in the relative position of 'also’, while there is a preference for close marking relationships. This article shifts attention to Hong Kong, where Cantonese and English converge, and 'also' finds counterparts in Cantonese ‘jaa’ and Mandarin ‘ye’. Employing a corpus-based data-driven method, we investigate the syntactic position of 'also' in both HK and GB English. The study aims to ascertain whether HK English exhibits a greater 'syntactic freedom,' allowing for a more distant marking relationship with 'also' compared to GB English. The analysis involves a random extraction of 500 samples from both HK and GB English from the GloWbE corpus, forming a dataset (N=1000). Exclusions are made for cases where 'also' functions as an additive conjunct or serves as a copulative adverb, as well as sentences lacking sufficient indication that 'also' functions as a focusing particle. The final dataset comprises 820 tokens, with 416 for GB and 404 for HK, annotated according to the focused constituent and the relative position of ‘also’. Frequency analysis reveals significant differences in the relative position of 'also' and marking relationships between HK and GB English. Regression analysis indicates a preference in HK English for a distant marking relationship between 'also' and its focused constituent. Notably, the subject and other constituents emerge as significant predictors of a distant position for 'also.' Together, these findings underscore the nuanced linguistic dynamics in HK English and contribute to our understanding of language contact. It suggests that future pedagogical practice should consider incorporating the syntactic variation within English varieties, facilitating leaners’ effective communication in diverse English-speaking environments and enhancing their intercultural communication competence.

Keywords: also, Cantonese, English, focus marker, frequency analysis, language contact, logistic regression analysis

Procedia PDF Downloads 40

Authors: P. Parthiban, Sonu Rajak, N. Kannan, R. Dhanalakshmi

Abstract:

This paper deals with the Vehicle Routing Problem (VRP) with environmental considerations which is called Pollution Routing Problem (PRP). The objective is to minimize the operational and environmental costs. It consists of routing a number of vehicles to serve a set of customers, and determining fuel consumption, driver wages and their speed on each route segment, while respecting the capacity constraints and time windows. In this context, we presented an Ant Colony Optimization (ACO) approach, combined with a Speed Optimization Algorithm (SOA) to solve the PRP. The proposed solution method consists of two stages. Stage one is to solve a Vehicle Routing Problem with Time Window (VRPTW) using ACO and in the second stage a SOA is run on the resulting VRPTW solutions. Given a vehicle route, the SOA consists of finding the optimal speed on each arc of the route in order to minimize an objective function comprising fuel consumption costs and driver wages. The proposed algorithm tested on benchmark problem, the preliminary results show that the proposed algorithm is able to provide good solutions.

Keywords: ant colony optimization, CO2 emissions, combinatorial optimization, speed optimization, vehicle routing

Procedia PDF Downloads 305

Authors: Zhong Ma, Zhuping Wang, Congxin Liu, Xiangzeng Liu

Abstract:

Satellite imagery classification is a challenging problem with many practical applications. In this paper, we designed a deep convolution neural network (DCNN) to classify the satellite imagery. The contributions of this paper are twofold — First, to cope with the large-scale variance in the satellite image, we introduced the inception module, which has multiple filters with different size at the same level, as the building block to build our DCNN model. Second, we proposed a genetic algorithm based method to efficiently search the best hyper-parameters of the DCNN in a large search space. The proposed method is evaluated on the benchmark database. The results of the proposed hyper-parameters search method show it will guide the search towards better regions of the parameter space. Based on the found hyper-parameters, we built our DCNN models, and evaluated its performance on satellite imagery classification, the results show the classification accuracy of proposed models outperform the state of the art method.

Keywords: satellite imagery classification, deep convolution network, genetic algorithm, hyper-parameter optimization

Procedia PDF Downloads 283

Authors: Shuo Liu

Abstract:

In Chinese text classification tasks, redundant words and phrases can interfere with the formation of extracted and analyzed text information, leading to a decrease in the accuracy of the classification model. To reduce irrelevant elements, extract and utilize text content information more efficiently and improve the accuracy of text classification models. In this paper, the text in the corpus is first extracted using the TextRank algorithm for abstraction, the words in the abstract are used as nodes to construct a text graph, and then the graph attention network (GAT) is used to complete the task of classifying the text. Testing on a Chinese dataset from the network, the classification accuracy was improved over the direct method of generating graph structures using text.

Keywords: Chinese natural language processing, text classification, abstract extraction, graph attention network

Procedia PDF Downloads 80

Authors: Asma Omri, Noureddine Benothman, Sofiane Sayahi, Fethi Tlili, Hichem Besbes

Abstract:

Cycling, a popular mode of transportation, can also be perilous due to cyclists' vulnerability to collisions with vehicles and obstacles. This paper presents an innovative cyclist safety system based on radar technology designed to offer real-time collision risk warnings to cyclists. The system incorporates a low-power radar sensor affixed to the bicycle and connected to a microcontroller. It leverages radar point cloud detections, a clustering algorithm, and a supervised classifier. These algorithms are optimized for efficiency to run on the TI’s AWR 1843 BOOST radar, utilizing a coarse classification approach distinguishing between cars, trucks, two-wheeled vehicles, and other objects. To enhance the performance of clustering techniques, we propose a 2-Level clustering approach. This approach builds on the state-of-the-art Density-based spatial clustering of applications with noise (DBSCAN). The objective is to first cluster objects based on their velocity, then refine the analysis by clustering based on position. The initial level identifies groups of objects with similar velocities and movement patterns. The subsequent level refines the analysis by considering the spatial distribution of these objects. The clusters obtained from the first level serve as input for the second level of clustering. Our proposed technique surpasses the classical DBSCAN algorithm in terms of geometrical metrics, including homogeneity, completeness, and V-score. Relevant cluster features are extracted and utilized to classify objects using an SVM classifier. Potential obstacles are identified based on their velocity and proximity to the cyclist. To optimize the system, we used the View of Delft dataset for hyperparameter selection and SVM classifier training. The system's performance was assessed using our collected dataset of radar point clouds synchronized with a camera on an Nvidia Jetson Nano board. The radar-based cyclist safety system is a practical solution that can be easily installed on any bicycle and connected to smartphones or other devices, offering real-time feedback and navigation assistance to cyclists. We conducted experiments to validate the system's feasibility, achieving an impressive 85% accuracy in the classification task. This system has the potential to significantly reduce the number of accidents involving cyclists and enhance their safety on the road.

Keywords: 2-level clustering, coarse classification, cyclist safety, warning system based on radar technology

Procedia PDF Downloads 66

Authors: Ruilian Zhao, Shukai Zhang, Yan Wang, Weiwei Wang

Abstract:

Repairing broken test cases is an expensive and challenging task in evolving software systems. Although an automated repair technique with intent preservation has been proposed, but it does not take into account the association between test repairs and assertions, leading to a large number of irrelevant candidates and decreasing the repair capability. This paper proposes an assertion-driven test repair approach. Furthermore, an intent-oriented priority criterion is raised to guide the repair candidate generation, making the repairs closer to the intent of the test. In more detail, repair targets are determined through post-dominance relations between assertions and the methods that directly cause compilation errors. Then, test repairs are generated from the target in a bottom-up way, guided by the intent-oriented priority criteria. Finally, the generated repair candidates are prioritized to match the original test intent. The approach is implemented and evaluated on the benchmark of 4 open-source programs and 91 broken test cases. The result shows that the approach can fix 89% (81/91) of broken test cases, which is more effective than the existing intentpreserved test repair approach, and our intent-oriented priority criteria work well.

Keywords: test repair, test intent, software test, test case evolution

Procedia PDF Downloads 112

Authors: Ashis Pradhan, Mohan P. Pradhan

Abstract:

Pattern Matching is useful for recognizing character in a digital image. OCR is one such technique which reads character from a digital image and recognizes them. Line segmentation is initially used for identifying character in an image and later refined by morphological operations like binarization, erosion, thinning, etc. The work discusses a recognition technique that defines a set of morphological operators based on its orientation in a character. These operators are further categorized into groups having similar shape but different orientation for efficient utilization of memory. Finally the characters are recognized in accordance with the occurrence of frequency in hierarchy of significant pattern of those morphological operators and by comparing them with the existing database of each character.

Keywords: binary image, morphological patterns, frequency count, priority, reduction data set and recognition

Procedia PDF Downloads 399

Authors: Jiajun Wang, Xiaoge Li

Abstract:

The purpose of the aspect-level sentiment analysis task is to identify the sentiment polarity of aspects in a sentence. Currently, most methods mainly focus on using neural networks and attention mechanisms to model the relationship between aspects and context, but they ignore the dependence of words in different ranges in the sentence, resulting in deviation when assigning relationship weight to other words other than aspect words. To solve these problems, we propose a new aspect-level sentiment analysis model that combines a multi-channel convolutional network and graph convolutional network (GCN). Firstly, the context and the degree of association between words are characterized by Long Short-Term Memory (LSTM) and self-attention mechanism. Besides, a multi-channel convolutional network is used to extract the features of words in different ranges. Finally, a convolutional graph network is used to associate the node information of the dependency tree structure. We conduct experiments on four benchmark datasets. The experimental results are compared with those of other models, which shows that our model is better and more effective.

Keywords: aspect-level sentiment analysis, attention, multi-channel convolution network, graph convolution network, dependency tree

Procedia PDF Downloads 196

Authors: Félix David Suárez Bonilla

Abstract:

A sentiment analysis service to detect polarity (positive, neural, and negative), based on transfer learning, was built using a Spanish version of BERT and applied to tweets written in Spanish. The dataset that was used consisted of 11975 reviews, which were extracted from Google Play using the google-play-scrapper package. The BETO trained model used: the AdamW optimizer, a batch size of 16, a learning rate of 2x10⁻⁵ and 10 epochs. The system was tested using tweets of three presidential pre-candidates from Costa Rica. The system was finally validated using human labeled examples, achieving an accuracy of 83.3%.

Keywords: NLP, transfer learning, BERT, sentiment analysis, social media, opinion mining

Procedia PDF Downloads 161

Authors: A. Harras, A. Tsuji, K. Terada

Abstract:

Tracking of moving people has gained a matter of great importance due to rapid technological advancements in the field of computer vision. The objective of this study is to design a motion based detection and tracking multiple walking pedestrians randomly in different directions. In our proposed method, Gaussian mixture model (GMM) is used to determine moving persons in image sequences. It reacts to changes that take place in the scene like different illumination; moving objects start and stop often, etc. Background noise in the scene is eliminated through applying morphological operations and the motions of tracked people which is determined by using the Kalman filter. The Kalman filter is applied to predict the tracked location in each frame and to determine the likelihood of each detection. We used a benchmark data set for the evaluation based on a side wall stationary camera. The actual scenes from the data set are taken on a street including up to eight people in front of the camera in different two scenes, the duration is 53 and 35 seconds, respectively. In the case of walking pedestrians in close proximity, the proposed method has achieved the detection ratio of 87%, and the tracking ratio is 77 % successfully. When they are deferred from each other, the detection ratio is increased to 90% and the tracking ratio is also increased to 79%.

Keywords: automatic detection, tracking, pedestrians, counting

Procedia PDF Downloads 242

Authors: Haiyan Wu, Ying Liu, Shaoyun Shi

Abstract:

Authorship attribution is to extract features to identify authors of anonymous documents. Many previous works on authorship attribution focus on statistical style features (e.g., sentence/word length), content features (e.g., frequent words, n-grams). Modeling these features by regression or some transparent machine learning methods gives a portrait of the authors' writing style. But these methods do not capture the syntactic (e.g., dependency relationship) or semantic (e.g., topics) information. In recent years, some researchers model syntactic trees or latent semantic information by neural networks. However, few works take them together. Besides, predictions by neural networks are difficult to explain, which is vital in authorship attribution tasks. In this paper, we not only utilize the statistical style and content features but also take advantage of both syntactic and semantic features. Different from an end-to-end neural model, feature selection and prediction are two steps in our method. An attentive n-gram network is utilized to select useful features, and logistic regression is applied to give prediction and understandable representation of writing style. Experiments show that our extracted features can improve the state-of-the-art methods on three benchmark datasets.

Keywords: authorship attribution, attention mechanism, syntactic feature, feature extraction

Procedia PDF Downloads 121

Authors: Thrivikraman Aswathi, S. Advaith

Abstract:

As there can be cases where the use of real data is somehow limited, such as when it is hard to get access to a large volume of real data, we need to go for synthetic data generation. This produces high-quality synthetic data while maintaining the statistical properties of a specific dataset. In the present work, a generative adversarial network (GAN) is trained to produce multivariate time series (MTS) data since the MTS is now being gathered more often in various real-world systems. Furthermore, the GAN-generated MTS data is fed into a transformer-based deep learning architecture that carries out the data categorization into predefined classes. Further, the model is evaluated across various distinct domains by generating corresponding MTS data.

Keywords: GAN, transformer, classification, multivariate time series

Procedia PDF Downloads 112

Authors: Xiao Zhou, Jianlin Cheng

Abstract:

A single amino acid mutation can have a significant impact on the stability of protein structure. Thus, the prediction of protein stability change induced by single site mutations is critical and useful for studying protein function and structure. Here, we presented a deep learning network with the dropout technique for predicting protein stability changes upon single amino acid substitution. While using only protein sequence as input, the overall prediction accuracy of the method on a standard benchmark is >85%, which is higher than existing sequence-based methods and is comparable to the methods that use not only protein sequence but also tertiary structure, pH value and temperature. The results demonstrate that deep learning is a promising technique for protein stability prediction. The good performance of this sequence-based method makes it a valuable tool for predicting the impact of mutations on most proteins whose experimental structures are not available. Both the downloadable software package and the user-friendly web server (DNpro) that implement the method for predicting protein stability changes induced by amino acid mutations are freely available for the community to use.

Keywords: bioinformatics, deep learning, protein stability prediction, biological data mining

Procedia PDF Downloads 445

Authors: Najmeh Abedzadeh, Matthew Jacobs

Abstract:

An intrusion detection system (IDS) is a software application that monitors malicious activities and generates alerts if any are detected. However, most network activities in IDS datasets are normal, and the relatively few numbers of attacks make the available data imbalanced. Consequently, cyber-attacks can hide inside a large number of normal activities, and machine learning algorithms have difficulty learning and classifying the data correctly. In this paper, a comprehensive literature review is conducted on different types of algorithms for both implementing the IDS and methods in correcting the imbalanced IDS dataset. The most famous algorithms are machine learning (ML), deep learning (DL), synthetic minority over-sampling technique (SMOTE), and reinforcement learning (RL). Most of the research use the CSE-CIC-IDS2017, CSE-CIC-IDS2018, and NSL-KDD datasets for evaluating their algorithms.

Keywords: IDS, imbalanced datasets, sampling algorithms, big data

Procedia PDF Downloads 300

Authors: A. Rana, M. Badri, D. Lopez Behar, O. Yee, H. Al Bqaei

Abstract:

The building sector is one of the largest emitters of greenhouse gases. However, as operation energy demands of this sector decrease with more effective energy policies and strategies, there is an urgent need to parallel focus on the growing proportion of embodied carbons. In this regard, benchmarks on embodied carbon of buildings can provide a point of reference to compare and improve the environmental performance of buildings for the stakeholders. Therefore, embodied carbon benchmarks can serve as a useful tool to address climate change challenges. This research utilizes the method to provide a knowledge roadmap of embodied carbon benchmarks development and implementation trends. Two main databases, Web of Science and Engineering Village, are considered for the study. The mapping was conducted with the help of VosViewer tool to provide information regarding: the critical research areas; most cited authors and publications; and countries with the highest publications. It is revealed that the role of benchmarks in energy policies is an emerging trend. In addition, the research highlighted that in policies, embodied carbon benchmarks are gaining importance at the material, whole building, and building portfolio levels. This research reveals direction for improvement and future research and of relevance to building industry professionals, policymakers, and researchers.

Keywords: buildings embodied carbon benchmark, methods, policy

Procedia PDF Downloads 158

Authors: Yu-Jia Jian, Emily Chia-Yu Su, Hui-Ling Hsu, Jian-Jhih Chen

Abstract:

Coronary artery is the major supplier of myocardial blood flow. When fat and cholesterol are deposit in the coronary arterial wall, narrowing and stenosis of the artery occurs, which may lead to myocardial ischemia and eventually infarction. According to the World Health Organization (WHO), estimated 740 million people have died of coronary heart disease in 2015. According to Statistics from Ministry of Health and Welfare in Taiwan, heart disease (except for hypertensive diseases) ranked the second among the top 10 causes of death from 2013 to 2016, and it still shows a growing trend. According to American Heart Association (AHA), the risk factors for coronary heart disease including: age (> 65 years), sex (men to women with 2:1 ratio), obesity, diabetes, hypertension, hyperlipidemia, smoking, family history, lack of exercise and more. We have collected a dataset of 421 patients from a hospital located in northern Taiwan who received coronary computed tomography (CT) angiography. There were 300 males (71.26%) and 121 females (28.74%), with age ranging from 24 to 92 years, and a mean age of 56.3 years. Prior to coronary CT angiography, basic data of the patients, including age, gender, obesity index (BMI), diastolic blood pressure, systolic blood pressure, diabetes, hypertension, hyperlipidemia, smoking, family history of coronary heart disease and exercise habits, were collected and used as input variables. The output variable of the prediction module is the degree of coronary artery stenosis. The output variable of the prediction module is the narrow constriction of the coronary artery. In this study, the dataset was randomly divided into 80% as training set and 20% as test set. Four machine learning algorithms, including logistic regression, stepwise regression, neural network and decision tree, were incorporated to generate prediction results. We used area under curve (AUC) / accuracy (Acc.) to compare the four models, the best model is neural network, followed by stepwise logistic regression, decision tree, and logistic regression, with 0.68 / 79 %, 0.68 / 74%, 0.65 / 78%, and 0.65 / 74%, respectively. Sensitivity of neural network was 27.3%, specificity was 90.8%, stepwise Logistic regression sensitivity was 18.2%, specificity was 92.3%, decision tree sensitivity was 13.6%, specificity was 100%, logistic regression sensitivity was 27.3%, specificity 89.2%. From the result of this study, we hope to improve the accuracy by improving the module parameters or other methods in the future and we hope to solve the problem of low sensitivity by adjusting the imbalanced proportion of positive and negative data.

Keywords: decision support, computed tomography, coronary artery, machine learning

Procedia PDF Downloads 217

Authors: Nay Myo Sandar

Abstract:

With the growth of technologies and applications, a large amount of data has been producing at increasing rate from various resources such as social media networks, sensor devices, and other information serving devices. This large collection of massive, complex and exponential growth of dataset is called big data. The traditional database systems cannot store and process such data due to large and complexity. Consequently, cloud computing is a potential solution for data storage and processing since it can provide a pool of resources for servers and storage. However, moving large amount of data to and from is a challenging issue since it can encounter a high latency due to large data size. With respect to big data movement problem, this paper reviews the literature of previous works, discusses about research issues, finds out approaches for dealing with big data movement problem.

Keywords: Big Data, Cloud Computing, Big Data Movement, Network Techniques

Procedia PDF Downloads 68