3830 research outputs found
Sort by
Sentiment Analysis of Amazon Electronic Product Reviews using Deep Learning
No full textDue to increased usage of technology (social media, online marketing, internet services) over the years, Sentiment Analysis which is analysis done to understand the opinion ex-pressed in a piece of text, has become a hot and trending topic in the world today. Sentiment Analysis opens the door to a plethora of intriguing applications in almost every possible domain, from politics to social media rant on trending topics, education, movies, product and service reviews. A thriving aspect of Sentiment Analysis is the sentiment polarity in customer reviews; companies receive criticism and commendation from customers to understand their views on different products and determine which products are more favourable than others. This project attempts to build two deep learning models: Convolutional Neural Network (CNN) & Long Short-Term Memory (LSTM) to automatically detect and classify sentiment polarity in Amazon Electronic review dataset. The raw text is processed into their respective word vector representation using GloVe Embeddings. Accuracy, Precision, Recall, and F1 Score are used to assess the selected models. Both
baseline models obtain 93% Accuracy. The results demonstrate that the models are able to accurately classify the reviews
Research Newsletter: Issue One
No full textThis edition of the newsletter highlights a call to collate all research activity in DBS and profiles some of our researchers. The newsletter also previews some of the upcoming events, including Research Day 2021 and calls for submissions to the next issue of the DBS Business Journal
An examination of social media use and the FA Women’s Super League: The challenges of going professional
No full textThis paper explores the use of social media, specifically Twitter, in two English football clubs during their first season under the professional structure of the WSL. The study uses data from the clubs’ Twitter accounts and interviews with key employees. A total of 1088 tweets were analysed. Four themes emerged: player availability; status; cooperation and social media strategy. The findings showed that while women’s football has made great strides, the increase use of social media by players has been not supported by proper media training. Furthermore, the increase in status has not been matched by better funding. It is important than the increase awareness surrounding women’s football is matched by an increase in player well-being as this will impact the standing of women’s football
The relationship between the Dark Tetrad traits and multicultural attitudes in Ireland
No full textA study was conducted to examine the relationship between the Dark Tetrad personality traits and attitudes towards multiculturalism in Ireland, to examine gender differences in the prevalence of the Dark Tetrad traits, and gender differences in attitudes towards multiculturalism. Data was collected via a quantitative survey comprised of demographic questions and two psychological scales: The Short Dark Tetrad (SD4: Paulhus, Buckells, Trapnell & Jones, 2014) and the Multicultural Attitude Scale (MAS; Breugelmans & Van de Vijver, 2004). The results showed that Machiavellianism and Psychopathy had a moderate negative relationship with attitudes towards multiculturalism. Results also showed significant gender differences in Narcissism, Psychopathy and Sadism, as well as significant gender differences in attitudes towards multiculturalism. The findings of this study suggest that more research on dark personality traits and multicultural attitudes is needed to further develop understanding of the motivations of individuals that express prejudicial attitudes and engage in prejudicial behaviours
Valuation of a blockchain-based virtual machine resolving interoperability
No full textNew technology can be disruptive and sometimes it is hard to evaluate its impact and its value. Blockchain technology is still nascent and keep evolving to pace never seen before. To evaluate its value, society out a value to these different projects. In this case, Ren project is about interoperability that connects all blockchains together. As before Ren project, each blockchain behave like intranet. Each blockchain is its own network. However, with Ren project implementation, blockchains can now proceed to transfer of value between blockchains. To understand Ren project real value, a study on the protocol is necessary and identify the elements that impact its valuation. Once the protocol has been understood and the variables are identified, would it be possible to predict its value in future with deep learning. As a timeseries issue Long Short-Term Memory model will be used to predict the value of this project. This is only the beginning as this project keep evolving but it is still possible to forecast its progress with the current information. Flexibility is required as new elements would need to be implemented as the project connects more and more blockchains
Quality communication: is there a best practice for all library publishing programs?
No full textCommunication with editors and journal managers in a library publishing program looks different across all institutions. Each library publishing program differs in the amount of staff and support they have, so the amount of time available to spend overseeing each journal publication and communicating with their editors also varies. Library publishers have the additional challenge of working with both traditional publications and bespoke or otherwise explorative publications with less defined measurements of success and quality. Student-run journals or faculty journals that do not publish on a specific schedule, for example, are important for library publishing programs to support, but they pose specific challenges when it comes to editor communication and discussions of quality. At Penn State, our solution is to publish with two levels for our journal publications, a “Supported” and “Imprint” level, which allows us to differentiate between these publication types. With these levels, new journals are able to move up in support after we discern their publishing quality and timeliness. Other library programs do not differentiate between these publications and support all of them in similar ways. However, neither framework defines how often communication with each editor should be made, and thus editors are often only communicated with when a problem arises on their end, rather than with consistent follow up and guidance from the library publisher. I would like to suggest that a method of consistent communication is necessary for all types of publications, even if the application looks different across library publishing programs with different levels of staff and support. This paper and the discussion at the IFLA SIG on Library Publishing highlights some important differences between library and commercial publishers, and identifies some questions that library publishers should be asking to help formulate a method and model for communicating with editors, which can be adapted to work for all types of library publisher
COVID-19 and the dynamics of digital transformation: Reshaping the fashion industry
No full textThe Covid-19 pandemic outbreak has had a major impact on the fashion industry. Fashion retail has suffered tremendously, and to survive, retailers have had to accelerate their digital transformation stages. This research study aims to explore and draw conclusions on the impact of Covid-19 on the fashion retail industry, as well as exploring how the fashion industry has adapted to cater to an online audience through digital transformation. Through the use of an exploratory thematic approach, the study contextualises leading concepts through a series of interviews of fashion designers in leading retail roles within the Irish market. Four established themes were identified, including; Fashion’s adaptation to digital transformation, Seamless integration of digital maturity, Innovative digital transformation trends among fashion retailers and fashion’s utilisation of social media. The results concluded that the fashion retailers are immersing themselves in digital transformation and innovation, and retailers are effectively starting to merge online channels with in-store shopping experience
The Determinants of Customer Satisfaction and Customer Retention in subscription based streaming services in Ireland
Get PDFThe main aim of this dissertation is to determine what are the main factors that lead to an increase in customer satisfaction and customer retention in subscription based video steaming service. In order to achieve this purpose, primary data was collected by using quantitative research method. A survey was used and 93 participants respond to a questionnaire of 37 questions related to quality service, customer satisfaction and customer retention. A non-probability method was used in this study due to time and costs limitations. Moreover, the data analysis was done by using Cronbach’s alpha to test internal consistency of the variables
and multiple regression to assess the linear relationship between service quality, customer satisfaction and customer retention. The findings of this research indicate that quality service dimensions influence customer satisfaction and customer retention. Price, for example, is the factor that most satisfies users while system accessibility and user experience are the ones that most affect customer retention
Stress, anxiety, coping among parents of children with ASD in Ireland during Covid-19 Pandemic
No full textThe aim of this cross-sectional study was to investigate relationships between stress, anxiety and coping among parents of an ASD child in Ireland, during the Covid-19 pandemic. Parents (N=61, 55 women, 6 men) completed an anonymous online survey including standardised Brief COPE and DASS-21 Scales. Results found most parents’ had high well-being and preferred ‘approach coping’ style during Covid-19 pandemic. Avoidance coping was significantly positively correlated with higher stress and anxiety, whereas a approach coping was unrelated to stress or anxiety. Coping preferences did not differ between employed/not-employed groups. Stress and anxiety level were uncorrelated with age. Extremely severe anxiety predicted higher approach coping. Conversely, extremely severe stress predicted higher avoidance coping. The high stress-avoidance coping relationship replicates previous research findings. Future qualitative research should explore reasons why avoidance coping is used among parents with highest stress, to inform development of pre-interventions to improve their wellbeing and adaptive approach coping