A Novel Machine Learning Approach for Sentiment Analysis on Twitter Incorporating the Universal Language Model Fine-Tuning and SVM

Twitter sentiment detectors (TSDs) provide a better solution to evaluate the quality of service and product than other traditional technologies. The classification accuracy and detection performance of TSDs, which are extremely reliant on the performance of the classification techniques, are used, and the quality of input features is provided. However, the time required is a big problem for the existing machine learning methods, which leads to a challenge for all enterprises that aim to transform their businesses to be processed by automated workflows. Deep learning techniques have been utilized in several real-world applications in different fields such as sentiment analysis. Deep learning approaches use different algorithms to obtain information from raw data such as texts or tweets and represent them in certain types of models. These models are used to infer information about new datasets that have not been modeled yet. We present a new effective method of sentiment analysis using deep learning architectures by combining the “universal language model fine-tuning” (ULMFiT) with support vector machine (SVM) to increase the detection efficiency and accuracy. The method introduces a new deep learning approach for Twitter sentiment analysis to detect the attitudes of people toward certain products based on their comments. The extensive results on three datasets illustrate that our model achieves the state-of-the-art results over all datasets. For example, the accuracy performance is 99.78% when it is applied on the Twitter US Airlines dataset.

Impacts of COVID-19 on Electric Vehicle Charging Behavior: Data Analytics, Visualization, and Clustering

COVID-19 pandemic has infected millions and led to a catastrophic loss of lives globally. It has also significantly disrupted the movement of people, businesses, and industries. Additionally, electric vehicle (EV) users have faced challenges in charging their vehicles in public charging locations where there is a risk of COVID-19 exposure. However, a case study of EV charging behavior and its impacts during the SARS-CoV-2 is not addressed in the existing literature. This paper investigates the impacts of COVID-19 on EV charging behavior by analyzing the charging activity during the pandemic using a dataset from a public charging facility in the USA. Data visualization of charging behavior alongside significant timelines of the pandemic was utilized for analysis. Moreover, a cluster analysis using k-means, hierarchical clustering, and Gaussian mixture models was performed to identify common groups of charging behavior based on the vehicle arrival and departure times. Although the number of vehicles using the charging station was reduced significantly due to lockdown restrictions, the charging activity started to pick up again since May 2021 due to an increase in vaccination and easing of public restrictions. However, the charging activity currently still remains around half of the activity pre-pandemic. A noticeable decline in charging session length and an increase in energy consumption can be observed as well. Clustering algorithms identified three groups of charging behavior during the pandemic and their analysis and performance comparison using internal validation measures were also presented.

Visual SLAM Based Spatial Recognition and Visualization Method for Mobile AR Systems

The simultaneous localization and mapping (SLAM) market is growing rapidly with advances in Machine Learning, Drones, and Augmented Reality (AR) technologies. However, due to the absence of an open source-based SLAM library for developing AR content, most SLAM researchers are required to conduct their own research and development to customize SLAM. In this paper, we propose an open source-based Mobile Markerless AR System by building our own pipeline based on Visual SLAM. To implement the Mobile AR System of this paper, we use ORB-SLAM3 and Unity Engine and experiment with running our system in a real environment and confirming it in the Unity Engine’s Mobile Viewer. Through this experimentation, we can verify that the Unity Engine and the SLAM System are tightly integrated and communicate smoothly. In addition, we expect to accelerate the growth of SLAM technology through this research.

Offensive-Language Detection on Multi-Semantic Fusion Based on Data Augmentation

The rapid identification of offensive language in social media is of great significance for preventing viral spread and reducing the spread of malicious information, such as cyberbullying and content related to self-harm. In existing research, the public datasets of offensive language are small; the label quality is uneven; and the performance of the pre-trained models is not satisfactory. To overcome these problems, we proposed a multi-semantic fusion model based on data augmentation (MSF). Data augmentation was carried out by back translation so that it reduced the impact of too-small datasets on performance. At the same time, we used a novel fusion mechanism that combines word-level semantic features and n-grams character features. The experimental results on the two datasets showed that the model proposed in this study can effectively extract the semantic information of offensive language and achieve state-of-the-art performance on both datasets.

A New Integrated Multi-Criteria Decision-Making Model for Resilient Supplier Selection

Unexpected worldwide disruptions brought various challenges to supply chain management thus manipulating the research direction towards resilience. Since the supplier is one of the important supply chain elements, the challenges can be overcome through resilient supplier selection. Supplier selection is a multi-criteria decision-making problem where several criteria are involved. In this study, GRA-BWM-TOPSIS was proposed to evaluate resilient suppliers. Seven resilience criteria which were Quality, Lead Time, Cost, Flexibility, Visibility, Responsiveness and Financial Stability have been proposed and five experts were selected to provide judgments for the selection process. By using the proposed method, the criteria importance levels were obtained using GRA and the criteria weights were computed using BWM, together with a consistency test. TOPSIS was applied to evaluate the suppliers’ performances. Through a case study in a food manufacturing company, 10 suppliers were evaluated and ranked. A validation process was carried out and the managerial implications were provided to ensure the effectiveness of the proposed model. GRA-BWM-TOPSIS is suitable for resilient supplier selection when there are uncertainties and incomplete data.

Model to Program and Blockchain Approaches for Business Processes and Workflows in Finance

Business process modeling and verification have become an essential way to control and assure organizational evolution. We overview the opportunities for the application of blockchain in Business Process Management and Modeling in Finance and we focus on in-depth analysis of claim process in insurance as a use case. We investigate the utilization of blockchain technology for model checking of Workflow, Business Processes to ensure consistency, integrity, and security in a dynamically changing business environment. We create a UML profile for the blockchain, then we combine it with a UML activity diagram followed by a verification using Petri nets to guarantee a distributed computing system and scalable with mutable data. Our paper creates a unified picture of the approaches towards business processes modeling used in the financial industry organized around the set of premises intending to develop a future research agenda for blockchain business process modeling, specifically for the financial industry domain.

3D Printing and Implementation of Digital Twins: Current Trends and Limitations

Fabricating objects with desired mechanical properties by utilizing 3D printing methods can be expensive and time-consuming, especially when based only on a trial-and-error test modus operandi. Digital twins (DT) can be proposed as a solution to understand, analyze and improve the fabricated item, service system or production line. However, the development of relevant DTs is still hampered by a number of factors, such as a lack of full understanding of the concept of DTs, their context and method of development. In addition, the connection between existing conventional systems and their data is under development. This work aims to summarize and review the current trends and limitations in DTs for additive manufacturing, in order to provide more insights for further research on DT systems.

New Business Models on Artificial Intelligence—The Case of the Optimization of a Blast Furnace in the Steel Industry by a Machine Learning Solution

This article took the case of the adoption of a Machine Learning (ML) solution in a steel manufacturing process through a platform provided by a Canadian startup, Canvass Analytics. The content of the paper includes a study around the state of the art of AI/ML adoption in steel manufacturing industries to optimize processes. The work aimed to highlight the opportunities that bring new business models based on AI/ML to improve processes in traditional industries. Methodologically, bibliographic research in the Scopus database was performed to establish the conceptual framework and the state of the art in the steel industry, then the case was presented and analyzed, to finally evaluate the impact of the new business model on the operation of the steel mill. The results of the case highlighted the way the innovative business model, based on a No-Code/Low-Code solution, achieved results in less time than conventional approaches of analytics solutions, and the way it is possible to democratize artificial intelligence and machine learning in traditional industrial environments. This work was focused on opportunities that arise around new business models linked to AI. In addition, the study looked into the framework of the adoption of AI/ML in a traditional industrial environment toward a smart manufacturing approach. The contribution of this article was the proposal of an innovative methodology to put AI/ML in the hands of process operators. It aimed to show how it was possible to achieve better results in a less complex and time-consuming adoption process. The work also highlighted the need for an important quantity of data from the process to approach this kind of solution.

On the Use of LoRaWAN for Mobile Internet of Things: The Impact of Mobility

A long-range wide-area network (LoRaWAN) targets both mobile and static Internet of Things (IoT) applications; it is suited to IoT applications, which require a large coverage area while consuming less power at a low data rate; it provides a solution for transferring data between IoT devices with a minimum cost in terms of power, at the expense of higher latency. LoRaWAN was designed for static low-power long-range networks. However, several IoT solution applications involve the use of mobility. Therefore, this study investigates the usage of LoRaWAN in the field of mobile Internet of Things applications such as bike rentals, fleet monitoring, and wildlife and animal tracking applications. Using the OMNeT++ simulator, two different well-known mobility models are used to investigate the influence of mobility on the performance of mobile LoRaWAN. The results show that intense LoRaWAN networks can operate under a high velocity and varying traffic load. It can be observed that the random waypoint model combination yields a better performance, but at the cost of higher collisions and energy consumption. As a consequence, the results suggest the reconsideration of mobile IoT solutions over LoRaWAN.

Design and Fabrication of Modified SMA-Connector Sensor for Detecting Water Adulteration in Honey and Natural Latex

A detection system for water adulteration in honey is proposed. It consists of a modified SMA-connector sensor and a vector network analyzer. A modified SMA-connector sensor is applied to measure complex relative permittivity, electrical conductivity, and phase constant of honey samples with the open-ended method. The system is tested in the frequency range of 0.5–4.0 GHz at the sample temperature of 25 °C. The relationships between the complex relative permittivity, electrical conductivity, the phase constant, and the honey samples with different concentrations (0–30%w/w) are determined. The experimental results show that the real part of the complex relative permittivity is significantly proportional in honey samples with adulteration of water in the range of 0–30%w/w. The frequency of 0.6 GHz is a suitable frequency for detection with a real part of complex relative permittivity as an indicator. The frequency of 3.74 GHz is an appropriate frequency for detection with electrical conductivity as in indicator while the frequency of 4.0 GHz is suitable for detection with phase constant as an indicator. In addition, the data are analyzed with regression analysis. This technique is also performed on natural latex samples to determine the dry rubber content. The frequency of 0.5 GHz is a suitable frequency with a real part of complex relative permittivity as an indicator while the frequency of 4.0 GHz is a suitable frequency with an imaginary part of complex relative permittivity, electrical conductivity, and phase constant as the indicators. The results demonstrate that it is possible to apply this technique to determine the dry rubber content in the natural latex samples as well.