Boletín que incluye información científica sobre el COVID-19, incluye artículos científicos, artículos preprint y revisiones sistemáticas actualizado al 14 de Mayo de 2020.

Boletín que incluye información científica sobre el COVID-19, incluye artículos científicos, artículos preprint y revisiones sistemáticas actualizado al 15 de Mayo de 2020.

Defending models against Natural Language Processing adversarial attacks is a challenge because of the discrete nature of the text dataset. However, given the variety of Natural Language Processing applications, it is important to make text processing models more robust and secure. This paper aims to develop techniques that will help text processing models such as BERT to combat adversarial samples that contain misspellings. These developed models are more robust than off the shelf spelling checkers.

"Math anxiety is commonly defined as a feeling of tension, apprehension, or fear that interferes with math performance" (Ashcraft, 2002, p. 181). Symptoms of math anxiety are reported by 33% of students by the time they reach the age of 15, possibly contributing to this workforce dilemma (Organisation for Economic Co-operation and Development, 2013). Many models and perspectives of math anxiety have been established including conceptualizing math anxiety as a function of working memory deficits, sociocultural conditioning, lack of reappraisal, and anxiety as a precursor to escape-maintained behavior. Math anxiety is more common in individuals with certain disabilities, such as developmental dyscalculia and deaf and hard of hearing. Hembree (1990) conducted a meta-analysis of research evaluating intervention effects on math anxiety and its moderators. They reported that cognitive-behavioral interventions were most effective at reducing math anxiety. While definitions of, models examining, and causes and co-occurrences of math anxiety are well examined in current literature, Hembree's (1990) meta- analysis remains the only such investigation for the past 30 years. The purpose of the present study is to conduct an updated meta-analysis based on previous research (Hembree, 1990) but focused on interventions in K-12 school settings. We identified 11 articles between the years of 1990-2020 that met our inclusion criteria. From those articles, we calculated an omnibus effect size, tested homogeneity, evaluated publication bias, explored moderating variables, and assessed methodological rigor. Our Q statistic indicated homogeneity; however, the forest plot and I2 indicated a small amount of heterogeneity. The asymmetric shape of the funnel plot may be indicative of publication bias. The omnibus effect size was g = 0.316. The results of our moderator analysis indicated that math anxiety interventions produce the best results when conducted in targeted small groups. Additionally, three studies were considered methodologically sound. Our findings support the use of school-based interventions to reduce math anxiety, especially when those interventions are implemented as a targeted, small group intervention.

Today, we are living in a data-driven world. Due to a surge in data generation, there is a need for efficient and accurate techniques to analyze data. One such kind of data which is needed to be analyzed are text reviews given for movies. Rather than classifying the reviews as positive or negative, we will classify the sentiment of the reviews on the scale of one to ten. In doing so, we will compare two recurrent neural network algorithms Long short term memory(LSTM) and Gated recurrent unit(GRU). The main objective of this study is to compare the accuracies of LSTM and GRU models. For training models, we collected data from two different sources. For filtering data, we used porter stemming and stop words. We coupled LSTM and GRU with the convolutional neural networks to increase the performance. After conducting experiments, we have observed that LSTM performed better in predicting border values. Whereas, GRU predicted every class equally. Overall GRU was able to predict multiclass text data of movie reviews slightly better than LSTM. GRU was computationally expansive when compared to LSTM.

Date: (2022-06-01) Level: Master thesis in Business Administration, 15 cr. Institution: School of Business, Society and Engineering, Mälardalen University. Authors: Mohammad Battha, (1993-06-10) Fadi Zina (1989-01-04) Title: The impact of online reviews and influencers on customers' purchasing intention. Supervisor:Stylianos Papaioannou Keywords: Influencers, Online reviews, Purchasing intention, Generation Y-Z. Research question: What is the impact of online reviews and influencers on the purchaseintentions of generations Y and Z of a given product or service? Purpose: Authors are aiming to investigate how influencers and online reviews can boost thepurchase intention of a brand. Hence, the authors scrutinize some of the influencers' andonline reviews' characteristics and analyze their impacts on purchasing intention from ageneration Y and Z customer perspective. The authors foresee expanding the literature andopening a door for further research in this narrow area of study. Methods: a deductive research method was applied through direct questionnaires to anumber of online consumers to test hypotheses in this research, and later the data wasanalyzed through regression and correlations tables. Conclusion: The online reviews by previous customers have a strong impact on customer’spurchase intentions. Also, the message carried by the product ambassadors such as bloggersand celebrities impacts the customers’ purchase intentions. The influencers carry the image ofa product and thus it is significant for the marketing brands to consider a careful selection ofbrand ambassadors as this impacts greatly on the future customer purchase intention.

The aim of this degree project is to gain a deeper understanding of how online reviews affect consumers' purchase intentions when purchasing high-involvement products. Exploring the differences of effect between two separate generations of consumers, Generation Z and Millennials. The age intervals have been 18-41, 18-25 for Generation Z consumers and 26-41 for Millennials. This research is utilizing a deductive approach paired with a quantitative research strategy. The primary data collection method has been through an internet-based self-completed questionnaire, whereas the data later on has been analyzed. The analysis has been conducted through the Statistical Package for Social Sciences (SPSS) program allowing for the researchers to perform a series of statistical analyses. From the empirical findings and analysis conducted, it was concluded that all of the hypotheses were disproven and rejected, showing no significant difference. Through SPSS a series of independent sample t-tests were conducted to extract two means for the two generations. The statistical output displayed a higher level of impact on Millennials consumers contrary to the hypothesis, predicting Generation Z’s purchase intentions being significantly more impacted. Through SPSS the researcher have been able to compare how each generation compare against each other in terms of all the variables making up purchase intention (Involvement, Argument Quality, Source Credibility, Information Usefulness), but also performed t-tests to see how their answers differentiate in each question as well. This study could provide marketers with very insightful knowledge about how online reviews affect consumers differently, allowing for adaptation and optimizations.

Research Question: To what extent do social media reviews’ effect on the purchase   intention and actual buying behaviour of consumers differ in the beauty industry in a cross-country comparison?     Purpose:   The purpose of this study is to explore to what extent consumers in the beauty industry perceive the importance of social media reviews and if it has direct or indirect effect on either their purchase intentions or buying behaviors or if not on both. The aim is then to contribute to limited research available and to see if this connection differs with the country particularly between Sweden and Sri Lanka.   Methodology:     A descriptive qualitative research strategy was approached conducting semi-structured interviews with a total of fourteen participants both from Sweden and Sri Lanka. Open-ended questions were asked and thematic analysis was used when analyzing the collected primary data.   Conclusion: Respondents from Sweden were more engaged with social media reviews and their purchase decisions were based mostly on social media reviews. Sri Lankan respondents were interested in negative reviews while Swedish respondents were highly affected and motivated by positive reviews. Even if young consumers in Sri Lanka were motivated to purchase beauty products because of social media reviews, almost all of the Sri Lankan respondents’ purchase decisions were highly impacted by family and friends compared to effects of social media reviews.

The status of contemporary Czech music reviews (also the Czech cultural criticism) ties in with mass culture and trends of postmodern thinking supports. Subjective journalist genres type reviews and critics look like the relevant expert testing about artistic performance, their real aim is economic profitability and fixing a position in their discourse domain. Mass culture can not offer an alternative to itselfa nd that's the reason, why central dichotomy variation - scheme is applicable not only to its products (music CDs, theatrical performances, a new book at he art), but also on the way, which is about these products further talks . Subjective meta - review has a the specific positron in the mass media.

"Math anxiety is commonly defined as a feeling of tension, apprehension, or fear that interferes with math performance" (Ashcraft, 2002, p. 181). Symptoms of math anxiety are reported by 33% of students by the time they reach the age of 15, possibly contributing to this workforce dilemma (Organisation for Economic Co-operation and Development, 2013). Many models and perspectives of math anxiety have been established including conceptualizing math anxiety as a function of working memory deficits, sociocultural conditioning, lack of reappraisal, and anxiety as a precursor to escape-maintained behavior. Math anxiety is more common in individuals with certain disabilities, such as developmental dyscalculia and deaf and hard of hearing. Hembree (1990) conducted a meta-analysis of research evaluating intervention effects on math anxiety and its moderators. They reported that cognitive-behavioral interventions were most effective at reducing math anxiety. While definitions of, models examining, and causes and co-occurrences of math anxiety are well examined in current literature, Hembree's (1990) meta- analysis remains the only such investigation for the past 30 years. The purpose of the present study is to conduct an updated meta-analysis based on previous research (Hembree, 1990) but focused on interventions in K-12 school settings. We identified 11 articles between the years of 1990-2020 that met our inclusion criteria. From those articles, we calculated an omnibus effect size, tested homogeneity, evaluated publication bias, explored moderating variables, and assessed methodological rigor. Our Q statistic indicated homogeneity; however, the forest plot and I2 indicated a small amount of heterogeneity. The asymmetric shape of the funnel plot may be indicative of publication bias. The omnibus effect size was g = 0.316. The results of our moderator analysis indicated that math anxiety interventions produce the best results when conducted in targeted small groups. Additionally, three studies were considered methodologically sound. Our findings support the use of school-based interventions to reduce math anxiety, especially when those interventions are implemented as a targeted, small group intervention.