Search results

Sarcasm is a linguistic expression that usually carries the opposite meaning of what is being said by words, thus making it difficult for machines to discover the actual meaning. It is mainly distinguished by the inflection with which it is spoken, with an undercurrent of irony, and is largely dependent on context, which makes it a difficult task for computational analysis. Moreover, sarcasm expresses negative sentiments using positive words, allowing it to easily confuse sentiment analysis models. This paper aims to demonstrate the task of sarcasm detection using the approach of machine and deep learning.

For the purpose of sarcasm detection, machine and deep learning models were used on a data set consisting of 1.3 million social media comments, including both sarcastic and non-sarcastic comments. The data set was pre-processed using natural language processing methods, and additional features were extracted and analysed. Several machine learning models, including logistic regression, ridge regression, linear support vector and support vector machines, along with two deep learning models based on bidirectional long short-term memory and one bidirectional encoder representations from transformers (BERT)-based model, were implemented, evaluated and compared.

The performance of machine and deep learning models was compared in the task of sarcasm detection, and possible ways of improvement were discussed. Deep learning models showed more promise, performance-wise, for this type of task. Specifically, a state-of-the-art model in natural language processing, namely, BERT-based model, outperformed other machine and deep learning models.

This study compared the performance of the various machine and deep learning models in the task of sarcasm detection using the data set of 1.3 million comments from social media.

Unlike traditional industries, the e-cigarette is an epoch-making innovative product originating in China and occupying an absolute competitive advantage in the international market. The traditional A-U model describes the laws and characteristics of technological innovation in developed countries. In contrast, the inverse A-U model depicts the process of “secondary innovation” in late-developing countries through digestion and absorption. This paper aims to find out that if the e-cigarette, as a “first innovation” industry in a late-developing country, conform to the A-U model or conform to the “inverse A-U model”.

This paper takes the patent data of e-cigarettes from 2004 to 2021 as the research object, and uses Python’s Jieba segment words to divide product innovation and process innovation, and then uses statistical analysis methods to conduct empirical analyses on these data.

Thus, an improved A-U model suitable for the e-cigarette industry is proposed. In this model, product innovation in the e-cigarette industry appeared earlier than process innovation, but the synchronous development of product and process innovation is not lagging. The improved A-U model in the e-cigarette industry is not only different from the traditional A-U model but also does not conform to the inverse A-U model.

It is conducive to expanding and clarifying the theoretical contribution and applicable boundaries of the A-U model and has sparked thinking and exploration of the A-U model in e-cigarettes and emerging industries.

On this basis, suggestions on the development path and countermeasures of the e-cigarette industry are put forward.

Based on the e-cigarette industry, this paper takes patents as the research object and provides the method of dividing product innovation and process innovation, and proposes an A-U model suitable for the e-cigarette industry on this basis. By comparing the traditional A-U model with the inverse A-U model in latecomer countries, the background and causes of e-cigarette A-U model heterogeneity are analyzed from different stages and overall morphology. Based on this, the heterogeneity characteristics of e-cigarette innovation are summarized and sorted out.

This study explores whether herd behavior exists for equity crowdfunding investors in China and whether this herding is rational.

Based on signaling theory and social learning theory, two hypotheses were proposed. This study employed two approaches to collect data. First, this paper analyzed 3,041 investments on an equity crowdfunding platform in China using Python programming and built a panel data model. Second, based on a unique experiment design, this study conducted several relevant herd behavior simulation experiments.

We found that investors in the Chinese equity crowdfunding market exhibit herd behavior and that this herding is rational. Project attributes play a negative role in moderating the relationship between the current investment amount and cumulative investments. Experimental results further support our findings.

This study contributes to the emerging literature on herding in crowdfunding by focusing on equity crowdfunding in China. We are the first to explore whether Chinese equity crowdfunding investors exhibit rational herding behavior. The study is also original in applying social learning theory to equity crowdfunding and in using both actual crowdfunding campaigns and experimental approaches to collect data. This study has valuable implications to practice.

The purpose of the paper is to conduct a numerical and experimental investigation into the properties of nanofluids containing spherical nanoparticles of random sizes flowing through a porous medium. The study aims to understand how the thermophysical properties of the nanofluid are affected by factors such as nanoparticle volume fraction, permeability of the porous medium, and pore size. The paper provides insights into the behavior of nanofluids in complex environments and explores the impact of varying conditions on key properties such as thermal conductivity, density, viscosity, and specific heat. Ultimately, the research contributes to the broader understanding of nanofluid dynamics and has potential implications for engineering and industrial applications in porous media.

This paper investigates nanofluids with spherical nanoparticles in a porous medium, exploring thermal conductivity, density, specific heat, and dynamic viscosity. Studying three compositions, the analysis employs the classical Maxwell model and Koo and Kleinstreuer’s approach for thermal conductivity, considering particle shape and temperature effects. Density and specific heat are defined based on mass and volume ratios. Dynamic viscosity models, including Brinkman’s and Gherasim et al.'s, are discussed. Numerical simulations, implemented in Python using the Langevin model, yield results processed in Origin Pro. This research enhances understanding of nanofluid behavior, contributing valuable insights to porous media applications.

This study involves a numerical examination of nanofluid properties, featuring spherical nanoparticles of varying sizes suspended in a base fluid with known density, flowing through a porous medium. Experimental findings reveal a notable increase in thermal conductivity, density, and viscosity as the volume fraction of particles rises. Conversely, specific heat experiences a decrease with higher particle volume concentration.xD; xA; The influence of permeability and pore size on particle volume fraction variation is a key focus. Interestingly, while the permeability of the medium has a significant effect, it is observed that it increases with permeability. This underscores the role of the medium’s nature in altering the thermophysical properties of nanofluids.

This paper presents a novel numerical study on nanofluids with randomly sized spherical nanoparticles flowing in a porous medium. It explores the impact of porous medium properties on nanofluid thermophysical characteristics, emphasizing the significance of permeability and pore size. The inclusion of random nanoparticle sizes adds practical relevance. Contrasting trends are observed, where thermal conductivity, density, and viscosity increase with particle volume fraction, while specific heat decreases. These findings offer valuable insights for engineering applications, providing a deeper understanding of nanofluid behavior in porous environments and guiding the design of efficient systems in various industrial contexts.

The Internet has changed consumer decision-making and influenced business behaviour. User-generated product information is abundant and readily available. This paper argues that user-generated content can be efficiently utilised for business intelligence using data science and develops an approach to demonstrate the methods and benefits of the different techniques.

Using Python Selenium, Beautiful Soup and various text mining approaches in R to access, retrieve and analyse user-generated content, we argue that (1) companies can extract information about the product attributes that matter most to consumers and (2) user-generated reviews enable the use of text mining results in combination with other demographic and statistical information (e.g. ratings) as an efficient input for competitive analysis.

The paper shows that combining different types of data (textual and numerical data) and applying and combining different methods can provide organisations with important business information and improve business performance.

The paper shows that combining different types of data (textual and numerical data) and applying and combining different methods can provide organisations with important business information and improve business performance.

The study makes several contributions to the marketing and management literature, mainly by illustrating the methodological advantages of text mining and accompanying statistical analysis, the different types of distilled information and their use in decision-making.

The present study distinguishes itself by pioneering an innovative framework that integrates key elements of prospect theory and the fundamental principles of electronic word of mouth (EWOM) to forecast Bitcoin/USD price fluctuations using Twitter sentiment analysis.

We utilized Twitter data as our primary data source. We meticulously collected a dataset consisting of over 3 million tweets spanning a nine-year period, from 2013 to 2022, covering a total of 3,215 days with an average daily tweet count of 1,000. The tweets were identified by utilizing the “bitcoin” and/or “btc” keywords through the snscrape python library. Diverging from conventional approaches, we introduce four distinct variables, encompassing normalized positive and negative sentiment scores as well as sentiment variance. These refinements markedly enhance sentiment analysis within the sphere of financial risk management.

Our findings highlight the substantial impact of negative sentiments in driving Bitcoin price declines, in contrast to the role of positive sentiments in facilitating price upswings. These results underscore the critical importance of continuous, real-time monitoring of negative sentiment shifts within the cryptocurrency market.

Our study holds substantial significance for both risk managers and investors, providing a crucial tool for well-informed decision-making in the cryptocurrency market. The implications drawn from our study hold notable relevance for financial risk management.

We present an innovative framework combining prospect theory and core principles of EWOM to predict Bitcoin price fluctuations through analysis of Twitter sentiment. Unlike conventional methods, we incorporate distinct positive and negative sentiment scores instead of relying solely on a single compound score. Notably, our pioneering sentiment analysis framework dissects sentiment into separate positive and negative components, advancing our comprehension of market sentiment dynamics. Furthermore, it equips financial institutions and investors with a more detailed and actionable insight into the risks associated not only with Bitcoin but also with other assets influenced by sentiment-driven market dynamics.

The purpose of this study is to address why managers enter the excessive market. A comparison of the facts and perceptions of entrants relative to success in the market shows that many entrants are confident about the viability of their businesses and enter the market. Accordingly, the authors simulate market entry decisions to detect behavioral biases.

The authors adapted the entry decisions simulation method, which is supported by the theoretical foundations of signal detection theory (SDT) and signaling theory. The simulation model is implemented on the Anaconda platform and written in Python 3.

The results of this study suggest that overestimation relates to excess market entry. Also, the proportion of excess entry under difficult conditions is always higher than under easy conditions.

This research helps managers and firms think about their and their competitors' abilities and evaluate them before entering the market. Policymakers and practitioners can also design programs such as experiential learning to help entrants assess their skills.

So far, no research has investigated the role of overconfidence under different market conditions. Accordingly, this study contributes to the current market entry literature by disentangling the debate between absolute and relative confidence and by considering the role of task difficulty.

It is well known in the mining industry that the increase in failures and breakdowns is due mainly to a poor maintenance policy for the equipment, in addition to the difficult access that specialized personnel have to combat the breakdown, which translates into more machine downtime. For this reason, this study aims to propose a remote assistance model for diagnosing and repairing critical breakdowns in mining industry trucks using augmented reality techniques and data analytics with a quality approach that considerably reduces response times, thus optimizing human resources.

In this work, the six-phase CRIPS-DM methodology is used. Initially, the problem of fault diagnosis in trucks used in the extraction of material in the mining industry is addressed. The authors then propose a model under study that seeks a real-time connection between a service technician attending the truck at the mine site and a specialist located at a remote location, considering the data transmission requirements and the machine's characterization.

It is considered that the theoretical results obtained in the development of this study are satisfactory from the business point of view since, in the first instance, it fulfills specific objectives related to the telecare process. On the other hand, from the data mining point of view, the results manage to comply with the theoretical aspects of the establishment of failure prediction models through the application of the CRISP-DM methodology. All of the above opens the possibility of developing prediction models through machine learning and establishing the best model for the objective of failure prediction.

The original contribution of this work is the proposal of the design of a remote assistance model for diagnosing and repairing critical failures in the mining industry, considering augmented reality and data analytics. Furthermore, the integration of remote assistance, the characterization of the CAEX, their maintenance information and the failure prediction models allow the establishment of a quality-based model since the database with which the learning machine will work is constantly updated.

Corporate philanthropy is increasingly a vital decision-making basis for consumers to purchase and establish relationships with enterprises. However, few studies have examined corporate philanthropy from the perspective of community evolution. To address this gap, this study aims to provide a more in-depth and holistic investigation of corporate philanthropy by examining the evolution of social media brand communities caused by corporate philanthropy and the characteristics of consumer interactive behavior.

Web crawlers developed by Python were employed to collect data of ERKE from Sina Weibo (the Chinese equivalent of Twitter). A total of 2,736 posts and 7,774 comments were collected and investigated using social network and sentiment tendency analyses.

The results showed that the evolution of the social media brand community presented a prominent three-stage characteristic influenced by corporate philanthropy. The findings not only support the benefits of corporate philanthropy but also show the possible disadvantages. Besides, this study further concluded the characteristics of consumer interactive behavior in the social media brand community.

This paper addresses an attractive and practical issue related to the impact of corporate philanthropy. Moreover, this study is one of the first studies to examine the impact of corporate philanthropy in the context of the social media brand community. The findings of this study will provide a valuable reference for community operations and practitioners of brands.

This paper aims to explore the consumer insights and ethical concerns surrounding the online payday loan services available in the Google Play Store. This research was conducted to compare whether the presence or absence of debt collection protection acts in a country creates differences in consumer experiences regarding the ethics of payday loan collection. Specifically, the study compares customers’ experiences in both the Indonesian and US markets.

Indonesia and the USA were chosen because they have very different regulatory structures for the payday loan industry. The data was scraped using Python from 27 payday loan apps on the Indonesian Play Store, resulting in a total of 244,697 reviews extracted from the Indonesian market. For the US market, 446,010 reviews were extracted from 14 payday loan apps. The data was further analyzed using NVIVO.

The results suggest that consumers of payday loans in Indonesia and the USA hold positive views about the benefits of payday loan apps, as revealed by the word frequency and word cloud analysis. Notably, customers in both countries did not express any negative sentiments regarding the unethical interest rate charged by the payday loan, contradicting what is commonly reported in academic literature. However, a distinct pattern of unethical conduct was observed in both countries concerning marketing communication and debt collection practices. In the Indonesian market, payday loan companies were found to engage in unethical debt collection activities. In the US market, payday lenders exhibited unethical behavior in their marketing communication, particularly through deceptive advertising that makes promises to consumers that are not delivered.

The study aims to provide evidence on the various experiences of customers in the presence and absence of debt collection regulations using a novel methodology and a large sample, which strengthens the results and conclusions of the study. The study also intends to inform policymakers, particularly the Indonesian government, about the need for specific laws to regulate the debt collection process and prevent unethical practices. Ultimately, the study is expected to protect the rights of consumers from a deceptive marketing communication or unethical debt collection practices in both the Indonesian and US markets.