Search results

The goal of this study is to investigate the predictive performance of the machine and deep learning methods in predicting the CAC 40 index and its 40 constituent prices of the French stock market during the COVID-19 pandemic. The study objective in forecasting the CAC 40 index is to analyze if the index and the individual prices will preserve the continuous increase they acquired at the beginning of the administration of vaccination and containment measures or if the negative effect of the pandemic will be reflected in the future.

The authors apply two machine and deep learning methods (KNN and LSTM) and compare their performances to ARIMA time series model. Two scenarios have been considered: optimistic (high values) and pessimistic (low values) and four periods are examined: the period before COVID-19 pandemic, the period during the COVID-19, and the period of vaccination and containment. The last period is divided into two sub-periods: the test period and the prediction period.

The authors found that the KNN method performed better than LSTM and ARIMA in forecasting the CAC 40 index for both scenarios. The authors also identified that the positive effect of vaccination and containment outweighs the negative effect of the pandemic, and the recovery pattern is not even among major companies in the stock market.

The study empirical results have valuable practical implications for companies in the stock market to respond to unexpected events such as COVID-19, improve operational efficiency and enhance long-term competitiveness. Companies in the transportation sector should consider additional investment in R&D on communication and information technology, accelerate their digital capabilities, at least in some parts of their businesses, develop plans for lights out factories and supply chains to keep pace with changing times, and even include big data resources. Additionally, they should also use a mix of financing sources and securities in order to diversify their capital structure, and not rely only on equity financing as their share prices are volatile and below the pre-pandemic level. Considering portfolio allocation, the transportation sector was severely affected by the pandemic. This displays that transportation equities fail to be a candidate as a good diversifier during the health crisis. However, the diversification would be worth it while including assets related to the banking and industrial sectors. On another strand, the instability of this period induced an informational asymmetry among investors. This pessimistic mood affected the assets' value and created a state of disequilibrium opening up more opportunities to benefit from potential arbitrage profits.

The impact of COVID-19 on stock markets is significant and affects investor behavior, who suffered amplified losses in a very short period of time. In this regard, correct and well-informed decision-making by investors and other market participants requires careful analysis and accurate prediction of the stock markets during the pandemic. However, few studies have been conducted in this area, and those studies have either concentrated on some specific stock markets or did not apply the powerful machine learning and deep learning techniques such as LSTM and KNN. To the best of our knowledge, no research has been conducted that used these techniques to assess and forecast the CAC 40 French stock market during the pandemic. This study tries to close this gap in the literature.

Aspect-based sentiment analysis (ASA) is a task of sentiment analysis that requires predicting aspect sentiment polarity for a given sentence. Many traditional techniques use graph-based mechanisms, which reduce prediction accuracy and introduce large amounts of noise. The other problem with graph-based mechanisms is that for some context words, the feelings change depending on the aspect, and therefore it is impossible to draw conclusions on their own. ASA is challenging because a given sentence can reveal complicated feelings about multiple aspects.

This research proposed an optimized attention-based DL model known as optimized aspect and self-attention aware long short-term memory for target-based semantic analysis (OAS-LSTM-TSA). The proposed model goes through three phases: preprocessing, aspect extraction and classification. Aspect extraction is done using a double-layered convolutional neural network (DL-CNN). The optimized aspect and self-attention embedded LSTM (OAS-LSTM) is used to classify aspect sentiment into three classes: positive, neutral and negative.

To detect and classify sentiment polarity of the aspect using the optimized aspect and self-attention embedded LSTM (OAS-LSTM) model. The results of the proposed method revealed that it achieves a high accuracy of 95.3 per cent for the restaurant dataset and 96.7 per cent for the laptop dataset.

The novelty of the research work is the addition of two effective attention layers in the network model, loss function reduction and accuracy enhancement, using a recent efficient optimization algorithm. The loss function in OAS-LSTM is minimized using the adaptive pelican optimization algorithm, thus increasing the accuracy rate. The performance of the proposed method is validated on four real-time datasets, Rest14, Lap14, Rest15 and Rest16, for various performance metrics.

This study aims to understand and a lay a foundation of how analytics has been used in depression management, this study conducts a systematic literature review using two techniques – text mining and manual review. The proposed methodology would aid researchers in identifying key concepts and research gaps, which in turn, will help them to establish the theoretical background supporting their empirical research objective.

This paper explores a hybrid methodology for literature review (HMLR), using text mining prior to systematic manual review.

The proposed rapid methodology is an effective tool to automate and speed up the process required to identify key and emerging concepts and research gaps in any specific research domain while conducting a systematic literature review. It assists in populating a research knowledge graph that does not reach all semantic depths of the examined domain yet provides some science-specific structure.

This study presents a new methodology for conducting a literature review for empirical research articles. This study has explored an “HMLR” that combines text mining and manual systematic literature review. Depending on the purpose of the research, these two techniques can be used in tandem to undertake a comprehensive literature review, by combining pieces of complex textual data together and revealing areas where research might be lacking.

Historical newspaper collections provide a wealth of information about the past. Although the digitization of these collections significantly improves their accessibility, a large portion of digitized historical newspaper collections, such as those of KBR, the Royal Library of Belgium, are not yet searchable at article-level. However, recent developments in AI-based research methods, such as document layout analysis, have the potential for further enriching the metadata to improve the searchability of these historical newspaper collections. This paper aims to discuss the aforementioned issue.

In this paper, the authors explore how existing computer vision and machine learning approaches can be used to improve access to digitized historical newspapers. To do this, the authors propose a workflow, using computer vision and machine learning approaches to (1) provide article-level access to digitized historical newspaper collections using document layout analysis, (2) extract specific types of articles (e.g. feuilletons – literary supplements from Le Peuple from 1938), (3) conduct image similarity analysis using (un)supervised classification methods and (4) perform named entity recognition (NER) to link the extracted information to open data.

The results show that the proposed workflow improves the accessibility and searchability of digitized historical newspapers, and also contributes to the building of corpora for digital humanities research. The AI-based methods enable automatic extraction of feuilletons, clustering of similar images and dynamic linking of related articles.

The proposed workflow enables automatic extraction of articles, including detection of a specific type of article, such as a feuilleton or literary supplement. This is particularly valuable for humanities researchers as it improves the searchability of these collections and enables corpora to be built around specific themes. Article-level access to, and improved searchability of, KBR's digitized newspapers are demonstrated through the online tool (https://tw06v072.ugent.be/kbr/).

This research compares consultants' and contractors' professional practices and perspectives on the use of specifications in the engineering of built environment projects.

The study initiates with a review of literature, toward realizing (1) the need for inclusion of specifications, throughout the projects' stages (design, tender, construction and handing over), as well as the (2) causes and effects of defective specifications in design and execution of engineered projects. Additionally (3) the study investigates the challenges and causes of specification variations to projects' design and execution. The systematic methodology adopted in this research utilizes frequency indicators, to rank and correlate the perspectives of 60 professionals, who are experienced in consulting and contracting of mega public and private projects.

The research considers the uncertainties and differences that could be present in engineering developments and use of design specifications for projects' execution and handing over. The two groups of experts' (consultants and contractors) perspectives are usually observed as conflicting, in practice. Practices comparison of their experiences with engineering projects' specifications usability provides unique empirical findings. The analysis and assessment of the experts' knowledge, this research addresses gaps identified in the literature, providing valuable insights.

The study provides in-depth recommendations and discusses possible trends and methods for enhancing specifications' usability in design engineering and execution of built environment projects. The analysis and recommendations present and assess experts' knowledge while bridge gaps identified from the literature.

Green building (GB) maintenance is increasingly accepted in the construction industry, so it can now be interpreted as an industry best practice for maintenance planning. However, the performance competency and design knowledge of the practice's building control instrument process can be affected by its evaluation and the information management of building information modelling (BIM)–based model checking (BMC). These maintenance-planning problems have not yet been investigated in instances such as the Grenfell Tower fire (14 June 2017, approximately 80 fatalities) in North Kensington, West London.

This study proposes a theoretical framework for analysing the existing conceptualisation of BIM tools and techniques based on a critical review of GB maintenance environments. These are currently employed on GB maintenance ecosystems embedded in project teams that can affect BMC practices in the automation system process. In order to better understand how BMC is implemented in GB ecosystem projects, a quantitative case study is conducted in the Malaysian public works department (Jabatan Kerja Raya (JKR)).

GB ecosystem projects were not as effective as planned due to safety awareness, design planning, inadequate track insulation, environmental (in) compatibility and inadequate building access management. Descriptive statistics and an ANOVA were applied to analyse the data. The study is reinforced by a process flow, which is transformed into a theoretical framework.

Industry practitioners can use the developed framework to diagnose BMC application issues and leverage the staff competency inherent in an ecosystem to plan GB maintenance environments successfully.