Today many companies exist and market their products and services on social medias, and therefore may receive reviews and thoughts from their end-users directly in these social medias. Reading every text by hand can be time-consuming, so by analysing the sentiment for all texts give the companies an overview how positive or negative the users are on a specific subject. Sentiment analysis is a feature that Beanloop AB is interested in implementing in their future projects and this thesis research problem was to investigate how deep learning could be used for this task. It was done by conducting an experiment with deep learning and neural networks. Several convolutional neural network models were implemented with different settings to find a combination of settings that gave the highest accuracy on the given test dataset. There were two different kind of models, one kind classifying positive and negative, and the second classified the previous two categories but also neutral. The training dataset and the test dataset contained data from two recommendation sites, www.reco.se and se.trustpilot.com. The final result shows that when classifying three categories (positive, negative and neutral) the models had problems to reach an accuracy at 85%, were only one model reached 80% accuracy as best on the test dataset. However, when only classifying two categories (positive and negative) the models showed very good results and reached almost 95% accuracy for every model.

People's engagements in social communities online have over recent years proven to be of great importance on others' opinions when purchasing products. Word-of-Mouth (WOM) has been around since ancient times and is a concept about consumers communicating with each other about their experience about services and products. Over the last decades research on Word-of-Mouth has increased when it was discovered how it has the power to influence consumers' purchase behaviours. The modern era of the introduction of the internet gave Word-of-Mouth a new concept called Electronic Word-of-Mouth (eWOM) that has given people the opportunity to share their negative and positive experiences online with others by leaving reviews. These reviews can be read by customers with access to the internet and could change their purchase intentions and therefore affect the company's sales and reputation. It has given the companies the opportunity to read about customers reviews and the chance to respond to the reviews by leaving comments online. This gave this study the opportunity to research if companies in Luleå are responding to negative reviews. As well, if the customers believe that their purchase intention is being affected by other users' reviews, and if the customers believe the companies can change the outcome of a review by responding to them.  In this study, quantitative research was used to collect data through surveys in Facebook groups with questions limited to Luleå. The questions were divided into three sections: Reading reviews, Responding to reviews, and Purchase intention. The study found that there was a relationship between the variables reading reviews, responding to reviews, and purchase intention. Previous literature has meant that people write more negative reviews when they have the ability to complain and write reviews through the Internet. However, for this study in Luleå, it was shown that this is not the case. It was found that few of the respondents gave negative reviews about the businesses in Luleå and that the businesses in Luleå did not respond to them. It was also found that respondents read reviews before buying, which affected their intention to buy, with negative reviews being more important than positive ones. People in Luleå consider negative reviews trustworthy and believe that businesses can change the outcome of negative reviews by responding to them. Therefore, it is of interest for local businesses in Luleå to review and respond to negative online reviews as they will impact their business.

No description available.

Purpose: The authors have found a common interest to investigate how the use and impact of reviews online can be applied within the subject of tourism. The purpose of the study is to investigate how consumers use reviews in their information search process. The purpose is also to investigate how and if consumers are influenced by online reviews regarding hotel stays. Method: The study is based on a qualitative method in the form of ten semi structured interviews as primary collection. The study has a support from secondary data consisting of a literature review in the chapter of previous research. Theory: The theoretical background is based on three theories. The theories are The PrePurchase-Information process, The Major Influences On Individual Travel Behaviour and the theory From Word-Of-Mouth to Electronic Word-Of-Mouth. Empiricism: The empiricism consists of a presentation of ten semi structured interviews with anonymous respondents regarding their own experiences and beliefs about the topic of the study. The chapter also includes an overview of the previous research about the topic to support and enhance the primary empirical interview material. Findings: The digital society has led to an increased online presence, critical thinking and interest on how consumers travel and what hotels they choose. Impact is constantly present in their contact with reviews and surroundings. The individuals own interpretation and perception of reality is based primarily on how they use and are influenced by reviews. In combination with critical thinking there is an insight that anyone, can write anything.

The dilemma of money laundering is an undeniable problem faced by many institutions all over the world. Due to its prevalence, the need for organisations to deal with the problem has become a global priority, regardless of the size of the financial institutions. Despite the compliance AML efforts undertaken by various institutions, AML compliance appears to be a daunting challenge, the question posed by this research is whether there is a need for independent anti-money laundering (AML) compliance reviews, given the invasive nature of money laundering in financial institutions. Secondly, the research addresses the question of who is well positioned to perform the independent AML compliance reviews. In addressing the second question, the research will discuss why internal audit is an invaluable resource in terms of risk management processes and the reasons why they are an integral part of the AML solution. / Dissertation (MPhil)--University of Pretoria, 2013. / gm2014 / Auditing / unrestricted

This thesis explores the transformations of evidence in health research synthesis studies – studies that bring together evidence from a number of research reports on the same/ similar topic. It argues that health research synthesis is a broad and intriguing field in a state of pre-formation, in spite of the fact that it may appear well established if equated with its exemplar method – the systematic review inclusive of meta-analysis. Transformations of evidence are processes by which pieces of evidence are modified from what they are in the primary study report into what is needed in the synthesis study while, supposedly, having their integrity fully preserved. Such processes have received no focused attention in the literature. Yet they are key to the validity and reliability of synthesis studies. This work begins to describe them and explore their frequency, scope and drivers. A ‘meta-scientific’ perspective is taken, where ‘meta-scientific’ is understood to include primarily ideas from the philosophy of science and methodological texts in health research, and, to a lesser extent, social studies of science and psychology of science thinking. A range of meta-scientific ideas on evidence and factors that shape it guide the analysis of processes of “data extraction” and “coding” during which much evidence is transformed. The core of the analysis involves the application of an extensive Analysis Framework to 17 highly heterogeneous research papers on cancer. Five non-standard ‘injunctions’ complement the Analysis Framework – for comprehensiveness, extensive multiple coding, extreme transparency, combination of critical appraisal and critique, and for first coding as close as possible to the original and then extending towards larger transformations. Findings suggest even lower credibility of the current overall model of health research synthesis than initially expected. Implications are discussed and a radical vision for the future proposed.

No description available.

Background Selective reporting of outcomes within a published study based on their nature or direction can result in systematic differences between reported and unreported data. Direct evidence of outcome reporting bias is limited to case reports. Objective To study empirically the nature of outcome reporting bias in randomised controlled trials (RCTs). Methods Three cohorts of RCTs were identified: PubMed-indexed RCTs published in December 2000; trial protocols approved by a Danish ethics committee from 1994-95; and trial protocols funded by a government agency in Canada from 1990-98. Data on reported and unreported outcomes were recorded from all trial publications and a survey of authors. An outcome was considered incompletely reported if insufficient data were presented for meta-analysis. Odds ratios relating the completeness of outcome reporting to statistical significance were calculated for each trial, and then pooled using a random effects meta-analysis. Protocols and publications were also reviewed for discrepancies in primary outcome reporting. Results 519 trials with 10,557 outcomes, 102 trials with 3613 outcomes, and 48 trials with 1390 outcomes were identified for the PubMed, ethics committee, and funding agency cohorts respectively. 22%-35% of outcomes per parallel group study were, on average, incompletely reported for meta-analysis. Fully reported outcomes had a two- to three-fold higher odds of being statistically significant compared to incompletely reported outcomes. The most common reasons given for omitting outcomes included a lack of clinical importance, lack of statistical significance, and space constraints. Major discrepancies between primary outcomes in protocols and publications were found in one half of trials. Discussion and conclusions The reporting of trial outcomes is frequently inadequate for meta-analysis; is biased to favour statistical significance; and is inconsistent with pre-specified protocol outcomes. Unacknowledged modifications to outcomes specified in trial protocols constitute scientific misconduct. Meta-analyses may therefore produce inflated and unreliable estimates of treatment effect.

Thesis (MDram)--Stellenbosch University, 2013. / ENGLISH ABSTRACT: This study examines the role and function of theatre critics and reviews in the industry at present. Characteristics of a good review are also examined and the role of the traditional review in the ‘new media’ is questioned. The ‘new media’ offer the public the opportunity to have access reviews to within a few hours of the first performance. This implies that theatre reviews become more newsworthy through their immediacy. “New media” have the added advantage that the content of reviews can be enhanced, and that the notorious space limitations of the ‘old media’ no longer impact on the copy volume. Within the modern media the style and content of reviews and opinions depend on the specific media form. In the ‘new media’ subjective personal opinions have become more prominent than well-grounded, ethically accountable journalism which used to be required in the older forms of media. The ‘new media’, on the other hand, present journalism with a whole new platform. If utilised in a controlled and dynamic manner, these new technological and electronic developments, can contribute enormously to the theatre review industry. In the concluding chapter a model is suggested according to the which the ‘new media’ can be used as a platform to maintain, develop and advance theatre – and specifically criticism – in South Africa. / AFRIKAANSE OPSOMMING: Hierdie studie ondersoek die rol en funksie van die teaterresensent en die resensie in die hedendaagse bedryf. Die eienskappe van ’n goeie resensie word ook ondersoek en die gebruik van die tradisionele resensie in die “nuwe media” word bevraagteken. Die “nuwe media” bied ’n geleentheid om resensies aan die publiek beskikbaar te stel, enkele ure ná die eerste vertoning. Die implikasie van hierdie element is dat ’n teaterresensie dus nuuswaardig word in sy onmiddellikheid. “Nuwe media” het die voordeel dat inhoudsverryking in resensies kan plaasvind en dat die beperkings van die ou media nie meer geld met betrekking tot die lengte van die werk nie. Styl en inhoudsaspekte van resensies of opinies in die “nuwe media” is kwessies waar die vorme van die verskillende media wel ter sprake kom. Binne die “nuwe media” is die persoonlike, subjektiewe opinies meer pertinent as deurgrondelike joernalistiek wat eties verantwoordbaar moet wees en wat van die ou media-vorme verwag word. Die “nuwe media” bied andersins ’n hele nuwe platform vir joernalistiek, en indien hierdie nuwe tegnologiese en elektroniese verwikkelinge op ’n gekontroleerde en dinamiese wyse ingespan kan word, kan dat dit ’n enorme bydrae tot die resensiewese lewer. In die laaste hoofstuk word ’n model voorgestel wat die “nuwe media” as platform gebruik om die behoud, ontwikkeling en bevordering van teater, en spesifiek teaterkritiek in Suid-Afrika, te bewerkstellig.

We live in a world where social media allows everyone to have a voice regardless of their expertise on any subject. With so many anonymous voices giving their opinions are the expert reviews of film critics no longer as useful? Some may believe there is a disconnect between what critics like and what people like. With this in mind, this research puts the usefulness of expert movie reviews and word of mouth against each other as can be seen through the need for cognition scale and social media affinity scale. / text