Fuzzy Ontology and LSTM-Based Text Mining: A Transportation Network Monitoring System for Assisting Travel

Design and Development of IoT and Deep Ensemble Learning Based Model for Disease Monitoring and Prediction

With the rapidly increasing reliance on advances in IoT, we persist towards pushing technology to new heights. From ordering food online to gene editing-based personalized healthcare, disruptive technologies like ML and AI continue to grow beyond our wildest dreams. Early detection and treatment through AI-assisted diagnostic models have outperformed human intelligence. In many cases, these tools can act upon the structured data containing probable symptoms, offer medication schedules based on the appropriate code related to diagnosis conventions, and predict adverse drug effects, if any, in accordance with medications. Utilizing AI and IoT in healthcare has facilitated innumerable benefits like minimizing cost, reducing hospital-obtained infections, decreasing mortality and morbidity etc. DL algorithms have opened up several frontiers by contributing towards healthcare opportunities through their ability to understand and learn from different levels of demonstration and generalization, which is significant in data analysis and interpretation. In contrast to ML which relies more on structured, labeled data and domain expertise to facilitate feature extractions, DL employs human-like cognitive abilities to extract hidden relationships and patterns from uncategorized data. Through the efficient application of DL techniques on the medical dataset, precise prediction, and classification of infectious/rare diseases, avoiding surgeries that can be preventable, minimization of over-dosage of harmful contrast agents for scans and biopsies can be reduced to a greater extent in future. Our study is focused on deploying ensemble deep learning algorithms and IoT devices to design and develop a diagnostic model that can effectively analyze medical Big Data and diagnose diseases by identifying abnormalities in early stages through medical images provided as input. This AI-assisted diagnostic model based on Ensemble Deep learning aims to be a valuable tool for healthcare systems and patients through its ability to diagnose diseases in the initial stages and present valuable insights to facilitate personalized treatment by aggregating the prediction of each base model and generating a final prediction.

Ontology Learning Applications of Knowledge Base Construction for Microelectronic Systems Information

Knowledge base construction (KBC) using AI has been one of the key goals of this highly popular technology since its emergence, as it helps to comprehend everything, including relations, around us. The construction of knowledge bases can summarize a piece of text in a machine-processable and understandable way. This can prove to be valuable and assistive to knowledge engineers. In this paper, we present the application of natural language processing in the construction of knowledge bases. We demonstrate how a trained bidirectional long short-term memory or bi-LSTM neural network model can be used to construct knowledge bases in accordance with the exact ISO26262 definitions as defined in the GENIAL! Basic Ontology. We provide the system with an electronic text document from the microelectronics domain and the system attempts to create a knowledge base from the available information in textual format. This information is then expressed in the form of graphs when queried by the user. This method of information retrieval presents the user with a much more technical and comprehensive understanding of an expert piece of text. This is achieved by applying the process of named entity recognition (NER) for knowledge extraction. This paper provides a result report of the current status of our knowledge construction process and knowledge base content, as well as describes our challenges and experiences.

Big data analysis of the impact of COVID-19 on digital game industrial sustainability in South Korea

The COVID-19 pandemic has greatly influenced the lifestyle and entertainment activities of the society that has significantly increased the growth rate of the gaming industry. While the studies on the game industry, one of the leading content industries, related to the pandemic has been done from various perspectives, little attention has been taken in regards to how the pandemic have impacted on the national digital game production and its industrial sustainability as a whole. Thus, this study was conducted to analyze the changes in the domestic game industry before and after the COVID-19 outbreak using the big data analysis of semantic network. This study aims to understand the growing trend in the gaming industry that can be helpful for the marketing and production of future games, as well as to guide the establishment of the public game policies in the game industry. The results showed that the COVID-19 pandemic positively decreased the public's worries and the government's restrictions towards gaming due to game addiction as a mental disease. However, its sudden change in the gamer's attitudes and the current gaming policies implied that for the sustainable development of the domestic game production, laws and regulations related to the game industry need to be reliable and planned on a long term basis since the industry is immensely large and is also related to several industries such as computing, programming, arts, and story contents. Accordingly, it is necessary to build an industrial ecology through which cluster complexes specializing in developing startups and small and medium-sized business can grow along with political support.

Abdominal Aortic Thrombus Segmentation in Postoperative Computed Tomography Angiography Images Using Bi-Directional Convolutional Long Short-Term Memory Architecture

Abdominal aortic aneurysm (AAA) is a fatal clinical condition with high mortality. Computed tomography angiography (CTA) imaging is the preferred minimally invasive modality for the long-term postoperative observation of AAA. Accurate segmentation of the thrombus region of interest (ROI) in a postoperative CTA image volume is essential for quantitative assessment and rapid clinical decision making by clinicians. Few investigators have proposed the adoption of convolutional neural networks (CNN). Although these methods demonstrated the potential of CNN architectures by automating the thrombus ROI segmentation, the segmentation performance can be further improved. The existing methods performed the segmentation process independently per 2D image and were incapable of using adjacent images, which could be useful for the robust segmentation of thrombus ROIs. In this work, we propose a thrombus ROI segmentation method to utilize not only the spatial features of a target image, but also the volumetric coherence available from adjacent images. We newly adopted a recurrent neural network, bi-directional convolutional long short-term memory (Bi-CLSTM) architecture, which can learn coherence between a sequence of data. This coherence learning capability can be useful for challenging situations, for example, when the target image exhibits inherent postoperative artifacts and noises, the inclusion of adjacent images would facilitate learning more robust features for thrombus ROI segmentation. We demonstrate the segmentation capability of our Bi-CLSTM-based method with a comparison of the existing 2D-based thrombus ROI segmentation counterpart as well as other established 2D- and 3D-based alternatives. Our comparison is based on a large-scale clinical dataset of 60 patient studies (i.e., 60 CTA image volumes). The results suggest the superior segmentation performance of our Bi-CLSTM-based method by achieving the highest scores of the evaluation metrics, e.g., our Bi-CLSTM results were 0.0331 higher on total overlap and 0.0331 lower on false negative when compared to 2D U-net++ as the second-best.

Improved genetic algorithm optimized LSTM model and its application in short-term traffic flow prediction

Considering that the road short-term traffic flow has strong time series correlation characteristics, a new long-term and short-term memory neural network (LSTM)-based prediction model optimized by the improved genetic algorithm (IGA) is proposed to improve the prediction accuracy of road traffic flow. Firstly, an improved genetic algorithm (IGA) is proposed by dynamically adjusting the mutation rate and crossover rate of standard GA. Secondly, the parameters of the LSTM, such as the number of hidden units, training times, gradient threshold and learning rate, are optimized by the IGA. Therefore, the optimal parameters are obtained. In the analysis stage, 5-min short-term traffic flow data are used to demonstrate the superiority of the proposed method over the existing neural network algorithms. Finally, the results show that the Root Mean Square Error achieved by the proposed algorithm is lower than that achieved by the other neural network methods in both the weekday and weekend data sets. This verifies that the algorithm can adapt well to different kinds of data and achieve higher prediction accuracy.

Unboxing Deep Learning Model of Food Delivery Service Reviews Using Explainable Artificial Intelligence (XAI) Technique

The demand for food delivery services (FDSs) during the COVID-19 crisis has been fuelled by consumers who prefer to order meals online and have it delivered to their door than to wait at a restaurant. Since many restaurants moved online and joined FDSs such as Uber Eats, Menulog, and Deliveroo, customer reviews on internet platforms have become a valuable source of information about a company’s performance. FDS organisations strive to collect customer complaints and effectively utilise the information to identify improvements needed to enhance customer satisfaction. However, only a few customer opinions are addressed because of the large amount of customer feedback data and lack of customer service consultants. Organisations can use artificial intelligence (AI) instead of relying on customer service experts and find solutions on their own to save money as opposed to reading each review. Based on the literature, deep learning (DL) methods have shown remarkable results in obtaining better accuracy when working with large datasets in other domains, but lack explainability in their model. Rapid research on explainable AI (XAI) to explain predictions made by opaque models looks promising but remains to be explored in the FDS domain. This study conducted a sentiment analysis by comparing simple and hybrid DL techniques (LSTM, Bi-LSTM, Bi-GRU-LSTM-CNN) in the FDS domain and explained the predictions using SHapley Additive exPlanations (SHAP) and Local Interpretable Model-Agnostic Explanations (LIME). The DL models were trained and tested on the customer review dataset extracted from the ProductReview website. Results showed that the LSTM, Bi-LSTM and Bi-GRU-LSTM-CNN models achieved an accuracy of 96.07%, 95.85% and 96.33%, respectively. The model should exhibit fewer false negatives because FDS organisations aim to identify and address each and every customer complaint. The LSTM model was chosen over the other two DL models, Bi-LSTM and Bi-GRU-LSTM-CNN, due to its lower rate of false negatives. XAI techniques, such as SHAP and LIME, revealed the feature contribution of the words used towards positive and negative sentiments, which were used to validate the model.

Research on Feature Extraction and Chinese Translation Method of Internet-of-Things English Terminology

Feature extraction and Chinese translation of Internet-of-Things English terms are the basis of many natural language processing. Its main purpose is to extract rich semantic information from unstructured texts to allow computers to further calculate and process them to meet different types of NLP-based tasks. However, most of the current methods use simple neural network models to count the word frequency or probability of words in the text, and it is difficult to accurately understand and translate IoT English terms. In response to this problem, this study proposes a neural network for feature extraction and Chinese translation of IoT English terms based on LSTM, which can not only correctly extract and translate IoT English vocabulary but also realize the feature correspondence between English and Chinese. The neural network proposed in this study has been tested and trained on multiple datasets, and it basically fulfills the requirements of feature translation and Chinese translation of Internet-of-Things terms in English and has great potential in the follow-up work.

Tracking traffic congestion and accidents using social media data: A case study of Shanghai

Traffic congestion and accidents take a toll on commuters' daily experiences and society. Locating the venues prone to congestion and accidents and capturing their perception by public members is invaluable for transport policy-makers. However, few previous methods consider user perception toward the accidents and congestion in finding and profiling the accident- and congestion-prone areas, leaving decision-makers unaware of the subsequent behavior responses and priorities of retrofitting measures. This study develops a framework to identify and characterize the accident- and congestion-prone areas heatedly discussed on social media. First, we use natural language processing and deep learning to detect the accident- and congestion-relevant Chinese microblogs posted on Sina Weibo, a Chinese social media platform. Then a modified Kernel Density Estimation method considering the sentiment of microblogs is employed to find the accident- and congestion-prone regions. The results show that the 'congestion-prone areas' discussed on social media are mainly distributed throughout the historical urban core and the Northwest of Pudong New Area, in reasonably good agreements with actual congestion records. In contrast, the 'accident-prone areas' are primarily found in locations with severe accidents. Finally, the above venues are characterized in spatio-temporal and semantic aspects to understand the nature of the incidents and assess the priority level for mitigation measures. The outcomes can provide a reference for traffic authorities to inform resource allocation and prioritize mitigation measures in future traffic management.

Rider weed deep residual network-based incremental model for text classification using multidimensional features and MapReduce

Increasing demands for information and the rapid growth of big data have dramatically increased the amount of textual data. In order to obtain useful text information, the classification of texts is considered an imperative task. Accordingly, this article will describe the development of a hybrid optimization algorithm for classifying text. Here, pre-processing was done using the stemming process and stop word removal. Additionally, we performed the extraction of imperative features and the selection of optimal features using the Tanimoto similarity, which estimates the similarity between features and selects the relevant features with higher feature selection accuracy. Following that, a deep residual network trained by the Adam algorithm was utilized for dynamic text classification. Dynamic learning was performed using the proposed Rider invasive weed optimization (RIWO)-based deep residual network along with fuzzy theory. The proposed RIWO algorithm combines invasive weed optimization (IWO) and the Rider optimization algorithm (ROA). These processes are carried out under the MapReduce framework. Our analysis revealed that the proposed RIWO-based deep residual network outperformed other techniques with the highest true positive rate (TPR) of 85%, true negative rate (TNR) of 94%, and accuracy of 88.7%.

Design Demand Trend Acquisition Method Based on Short Text Mining of User Comments in Shopping Websites

In order to facilitate designers to explore the market demand trend of laptops and to establish a better “network users-market feedback mechanism”, we propose a design and research method of a short text mining tool based on the K-means clustering algorithm and Kano mode. An improved short text clustering algorithm is used to extract the design elements of laptops. Based on the traditional questionnaire, we extract the user’s attention factors, score the emotional tendency, and analyze the user’s needs based on the Kano model. Then, we select 10 laptops, process them by the improved algorithm, cluster the evaluation words and quantify the emotional orientation matching. Based on the obtained data, we design a visual interaction logic and usability test. These prove that the proposed method is feasible and effective.

Stock Price Movement Prediction Using Sentiment Analysis and CandleStick Chart Representation

Determining the price movement of stocks is a challenging problem to solve because of factors such as industry performance, economic variables, investor sentiment, company news, company performance, and social media sentiment. People can predict the price movement of stocks by applying machine learning algorithms on information contained in historical data, stock candlestick-chart data, and social-media data. However, it is hard to predict stock movement based on a single classifier. In this study, we proposed a multichannel collaborative network by incorporating candlestick-chart and social-media data for stock trend predictions. We first extracted the social media sentiment features using the Natural Language Toolkit and sentiment analysis data from Twitter. We then transformed the stock’s historical time series data into a candlestick chart to elucidate patterns in the stock’s movement. Finally, we integrated the stock’s sentiment features and its candlestick chart to predict the stock price movement over 4-, 6-, 8-, and 10-day time periods. Our collaborative network consisted of two branches: the first branch contained a one-dimensional convolutional neural network (CNN) performing sentiment classification. The second branch included a two-dimensional (2D) CNN performing image classifications based on 2D candlestick chart data. We evaluated our model for five high-demand stocks (Apple, Tesla, IBM, Amazon, and Google) and determined that our collaborative network achieved promising results and compared favorably against single-network models using either sentiment data or candlestick charts alone. The proposed method obtained the most favorable performance with 75.38% accuracy for Apple stock. We also found that the stock price prediction achieved more favorable performance over longer periods of time compared with shorter periods of time.

A Knowledge-Based Sense Disambiguation Method to Semantically Enhanced NL Question for Restricted Domain

Within the space of question answering (QA) systems, the most critical module to improve overall performance is question analysis processing. Extracting the lexical semantic of a Natural Language (NL) question presents challenges at syntactic and semantic levels for most QA systems. This is due to the difference between the words posed by a user and the terms presently stored in the knowledge bases. Many studies have achieved encouraging results in lexical semantic resolution on the topic of word sense disambiguation (WSD), and several other works consider these challenges in the context of QA applications. Additionally, few scholars have examined the role of WSD in returning potential answers corresponding to particular questions. However, natural language processing (NLP) is still facing several challenges to determine the precise meaning of various ambiguities. Therefore, the motivation of this work is to propose a novel knowledge-based sense disambiguation (KSD) method for resolving the problem of lexical ambiguity associated with questions posed in QA systems. The major contribution is the proposed innovative method, which incorporates multiple knowledge sources. This includes the question’s metadata (date/GPS), context knowledge, and domain ontology into a shallow NLP. The proposed KSD method is developed into a unique tool for a mobile QA application that aims to determine the intended meaning of questions expressed by pilgrims. The experimental results reveal that our method obtained comparable and better accuracy performance than the baselines in the context of the pilgrimage domain.

Scenario-Mining for Level 4 Automated Vehicle Safety Assessment from Real Accident Situations in Urban Areas Using a Natural Language Process

As the research and development activities of automated vehicles have been active in recent years, developing test scenarios and methods has become necessary to evaluate and ensure their safety. Based on the current context, this study developed an automated vehicle test scenario derivation methodology using traffic accident data and a natural language processing technique. The natural language processing technique-based test scenario mining methodology generated 16 functional test scenarios for urban arterials and 38 scenarios for intersections in urban areas. The proposed methodology was validated by determining the number of traffic accident records that can be explained by the resulting test scenarios. That is, the resulting test scenarios are valid and represent a matching rate between the test scenarios and the increased number of traffic accident records. The resulting functional scenarios generated by the proposed methodology account for 43.69% and 27.63% of the actual traffic accidents for urban arterial and intersection scenarios, respectively.

Traffic accident detection and condition analysis based on social networking data

Accurate detection of traffic accidents as well as condition analysis are essential to effectively restoring traffic flow and reducing serious injuries and fatalities. This goal can be obtained using an advanced data classification model with a rich source of traffic information. Several systems based on sensors and social networking platforms have been presented recently to detect traffic events and monitor traffic conditions. However, sensor-based systems provide limited information, and may fail owing to the long detection times and high false-alarm rates. In addition, social networking data are unstructured, unpredictable, and contain idioms, jargon, and dynamic topics. The machine learning algorithms utilized for traffic event detection might not extract valuable information from social networking data. In this paper, a social network-based, real-time monitoring framework is proposed for traffic accident detection and condition analysis using ontology and latent Dirichlet allocation (OLDA) and bidirectional long short-term memory (Bi-LSTM). First, the query-based search engine effectively collects traffic information from social networks, and the data preprocessing module transforms it into structured form. Second, the proposed OLDA-based topic modeling method automatically labels each sentence (e.g., traffic or non-traffic) to identify the exact traffic information. In addition, the ontology-based event recognition approach detects traffic events from traffic-related data. Next, the sentiment analysis technique identifies the polarity of traffic events employing user's opinions, which helps determine accurate conditions of traffic events. Finally, the FastText model and Bi-LSTM with softmax regression are trained for traffic event detection and condition analysis. The proposed framework is evaluated using traffic-related data, comparing OLDA and Bi-LSTM with existing topic modeling methods and traditional classifiers using word embedding models, respectively. Our system outperforms state-of-the-art methods and achieves accuracy of 97 %. This finding demonstrates that the proposed system is more efficient for traffic event detection and condition analysis, in comparison to other existing systems.

Graph Adaptation Network with Domain-Specific Word Alignment for Cross-Domain Relation Extraction

Cross-domain relation extraction has become an essential approach when target domain lacking labeled data. Most existing works adapted relation extraction models from the source domain to target domain through aligning sequential features, but failed to transfer non-local and non-sequential features such as word co-occurrence which are also critical for cross-domain relation extraction. To address this issue, in this paper, we propose a novel tripartite graph architecture to adapt non-local features when there is no labeled data in the target domain. The graph uses domain words as nodes to model the co-occurrence relation between domain-specific words and domain-independent words. Through graph convolutions on the tripartite graph, the information of domain-specific words is propagated so that the word representation can be fine-tuned to align domain-specific features. In addition, unlike the traditional graph structure, the weights of edges innovatively combine fixed weight and dynamic weight, to capture the global non-local features and avoid introducing noise to word representation. Experiments on three domains of ACE2005 datasets show that our method outperforms the state-of-the-art models by a big margin.

Using BiLSTM Networks for Context-Aware Deep Sensitivity Labelling on Conversational Data

Information privacy is a critical design feature for any exchange system, with privacy-preserving applications requiring, most of the time, the identification and labelling of sensitive information. However, privacy and the concept of “sensitive information” are extremely elusive terms, as they are heavily dependent upon the context they are conveyed in. To accommodate such specificity, we first introduce a taxonomy of four context classes to categorise relationships of terms with their textual surroundings by meaning, interaction, precedence, and preference. We then propose a predictive context-aware model based on a Bidirectional Long Short Term Memory network with Conditional Random Fields (BiLSTM + CRF) to identify and label sensitive information in conversational data (multi-class sensitivity labelling). We train our model on a synthetic annotated dataset of real-world conversational data categorised in 13 sensitivity classes that we derive from the P3P standard. We parameterise and run a series of experiments featuring word and character embeddings and introduce a set of auxiliary features to improve model performance. Our results demonstrate that the BiLSTM + CRF model architecture with BERT embeddings and WordShape features is the most effective (F1 score 96.73%). Evaluation of the model is conducted under both temporal and semantic contexts, achieving a 76.33% F1 score on unseen data and outperforms Google’s Data Loss Prevention (DLP) system on sensitivity labelling tasks.

Recognizing Human Daily Activity Using Social Media Sensors and Deep Learning

The human daily activity category represents individual lifestyle and pattern, such as sports and shopping, which reflect personal habits, lifestyle, and preferences and are of great value for human health and many other application fields. Currently, compared to questionnaires, social media as a sensor provides low-cost and easy-to-access data sources, providing new opportunities for obtaining human daily activity category data. However, there are still some challenges to accurately recognizing posts because existing studies ignore contextual information or word order in posts and remain unsatisfactory for capturing the activity semantics of words. To address this problem, we propose a general model for recognizing the human activity category based on deep learning. This model not only describes how to extract a sequence of higher-level word phrase representations in posts based on the deep learning sequence model but also how to integrate temporal information and external knowledge to capture the activity semantics in posts. Considering that no benchmark dataset is available in such studies, we built a dataset that was used for training and evaluating the model. The experimental results show that the proposed model significantly improves the accuracy of recognizing the human activity category compared with traditional classification methods.