This bachelor thesis quantitatively researches every tweet Donald Trump has written in february 2017 from his personal Twitter account @realDonaldTrump. This particular period was chosen because it was the first full month that Trump was president of the United States of America. That's why it would be interesting to examine his tweets in his new official role as president. The purpose of this study is to find out how Donald Trump used language and his content on Twitter during his first month as president in 2017. What time during the day Donald Trump published his tweets and how his tweets were received by his readers is also of interest to this study. To do this the study uses theories about power as an interactionfield, representation and encoding and decoding. This study relies on a quantitative content analasis as a method and through it Trumps tweets were gathered, compiled and then given a main and a secondary theme. The themes were negative, positive, neutral, political, personal and other. Besides the main and secondary themes, i counted the number of syllables and words Trump used and also how much response he got on the tweets in the form av replies, retweets and likes. The result shows that Trumps negative and political tweets got the most response, mostly in the form of likes. Words that Trump writes more than others are words like fake, news, and media. the result also showed that most of the words Trump writes on Twitter only contains one or two syllables. / Denna studie har kvantitativt undersökt alla Donald Trumps tweets från februari 2017. Dessa tweets härstammar från Trumps personliga twitterkonto @realDonaldTrump. Perioden valdes på grund av det var den första hela månaden som Donald Trump var USA:s president. Därför var det intressant att se hur Trump twittrade som president. Själva syftet med studien är att utifrån teorier om makt som interaktionsfält, representation och encoding och decoding ta reda på hur Donald Trump språkligt och innehållsligt använde Twitter under den första månaden som president 2017. Av intresse är också vilken tid på dygnet Donald Trump twittrar och hur hans tweets har mottagits av läsarna. Studien utgår från en kvantitativ innehållsanalys som metod och genom den har Trumps tweets sammanställts och kategoriserats upp i huvudteman och underteman, bestående av negativa, positiva, neutrala, politiska, personliga och annat. Utöver huvudteman och underteman räknades även antalet stavelser och ord Trump skrev. Studien analyserade bland annat hur mycket respons Trump fick på sina tweets och responsen han får är i form av reply, retweet och like. Resultatet tyder på att Trumps negativa och politiska tweets får mest respons, främst i form av likes. Det har även undersökts vilka ord Trump skriver mest på Twitter och hur många stavelser de innehåller och i resultatet kom det fram att ord som Trump skriver mycket är fake, news, och media med mera. Trump skriver även majoriteten av sina ord med en eller två stavelser.

Online communities provide a fertile ground for analyzing people\'s behavior and improving our understanding of social processes. For instance, when modeling social interaction online, it is important to understand when people are reacting to each other. Also, since both people and communities change over time, we argue that analyses of online communities that take time into account will lead to deeper and more accurate results. In many cases, however, users behavior can be easily missed: users react to content in many more ways than observed by explicit indicators (such as likes on Facebook or replies on Twitter) and poorly aggregated temporal data might hide, misrepresent and even lead to wrong conclusions about how users are evolving. In order to address the problem of detecting non-explicit responses, we present a new approach that uses tf-idf similarity between a user\'s own tweets and recent tweets by people they follow. Based on a month\'s worth of posting data from 449 ego networks in Twitter, this method demonstrates that it is likely that at least 11% of reactions are not captured by the explicit reply and retweet mechanisms. Further, these uncaptured reactions are not evenly distributed between users: some users, who create replies and retweets without using the official interface mechanisms, are much more responsive to followees than they appear. This suggests that detecting non-explicit responses is an important consideration in mitigating biases and building more accurate models when using these markers to study social interaction and information diffusion. We also address the problem of users evolution in Reddit based on comment and submission data from 2007 to 2014. Even using one of the simplest temporal differences between usersyearly cohortswe find wide differences in people\'s behavior, including comment activity, effort, and survival. Furthermore, not accounting for time can lead us to misinterpret important phenomena. For instance, we observe that average comment length decreases over any fixed period of time, but comment length in each cohort of users steadily increases during the same period after an abrupt initial drop, an example of Simpson\'s Paradox. Dividing cohorts into sub-cohorts based on the survival time in the community provides further insights; in particular, longer-lived users start at a higher activity level and make more and shorter comments than those who leave earlier. These findings both give more insight into user evolution in Reddit in particular, and raise a number of interesting questions around studying online behavior going forward. / Comunidades online proporcionam um ambiente fértil para análise do comportamento de indivíduos e processos sociais. Por exemplo, ao modelarmos interações sociais online, é importante compreendemos quando indivíduos estão reagindo a outros indivíduos. Além disso, pessoas e comunidades mudam com o passar do tempo, e levar em consideração sua evolução temporal nos leva a resultados mais precisos. Entretanto, em muitos casos, o comportamento dos usuários pode ser perdido: suas reações ao conteúdo ao qual são expostos não são capturadas por indicadores explícitos (likes no Facebook, replies no Twitter). Agregações temporais de dados pouco criteriosas podem ocultar, enviesar ou até levar a conclusões equivocadas sobre como usuários evoluem. Apresentamos uma nova abordagem para o problema de detectar respostas não-explicitas que utiliza similaridade tf-idf entre tweets de um usuário e tweets recentes que este usuário recebeu de quem segue. Com base em dados de postagens de um mês para 449 redes egocêntricas do Twitter, este método evidencia que temos um volume de ao menos 11% de reações não capturadas pelos mecanismos explicitos de reply e retweet. Além disso, essas reações não capturadas não estão uniformemente distribuídas entre os usuários: alguns usuários que criam replies e retweets sem utilizar os mecanismos formais da interface são muito mais responsivos a quem eles seguem do que aparentam. Isso sugere que detectar respostas não-explicitas é importante para mitigar viéses e construir modelos mais precisos a fim de estudar interações sociais e difusão de informação. Abordamos o problema de evolução de usuários no Reddit com base em dados entre o período de 2007 a 2014. Utilizando métodos simples de diferenciação temporal dos usuários -- cohorts anuais -- encontramos amplas diferenças entre o comportamento, que incluem criação de comentários, métricas de esforço e sobrevivência. Desconsiderar a evolução temporal pode levar a equívocos a respeito de fenômenos importantes. Por exemplo, o tamanho médio dos comentários na rede decresce ao longo de qualquer intervalo de tempo, mas este tamanho é crescente em cada uma das cohorts de usuários no mesmo período, salvo de uma queda inicial. Esta é uma observação do Paradoxo de Simpson. Dividir as cohorts de usuários em sub-cohorts baseadas em anos de sobrevivência na rede nos fornece uma perspectiva melhor; usuários que sobrevivem por mais tempo apresentam um maior nível de atividade inicial, com comentários mais curtos do que aqueles que sobrevivem menos. Com isto, compreendemos melhor como usuários evoluem no Reddit e levantamos uma série de questões a respeito de futuros desdobramentos do estudo de comportamento online.

Research on bot detection on online social networks has received a considerable amount of attention in Swedish news media. Recently however, criticism of the research field of bot detection on onlinesocial networks has been presented, highlighting the need to investigate the research field to determine if information based on flawed research has been spread. To investigate the research field, this study has attempted to review the process of bot detection on online social networks and evaluate the proposed criticism of current bot detection research by: conducting a literature review of bots on online social networks, conducting a literature review of methods for bot detection on online social networks, and detecting bots in three different politically associated data sets with Swedish Twitter accounts usingfive different bot detection methods. Results of the study showed minor evidence that previous research may have been flawed. Still, based on the literature review of bot detection methods, it was determined that this criticism was not extensive enough to critique the research fieldof bot detection on online social networks as a whole. Further, problems highlighted in the criticism were recognized to potentially have arose from a lack of differentiation between bot types in research. An insufficient differentiation between bot types in research was also acknowledged as a factor which could lead to difficulties ingeneralizing the results from bot detection studies measuring the effect of bots on political opinions. Instead, the study acknowledged that a good bot differentiation could potentially improve bot detection.

World health organization (WHO) har listat tio hot mot världshälsan varav två av dessa är pandemier och vaccinskepticism. Enligt forskare kommer pandemier drabba mänskligheten samtidigt som vaccinskepticismen ökar. Individer som är kritiska till vaccin tillhör antivaccinationsrörelsen och de menar på att vaccin är ett sätt att skada befolkningen. Antivaccinationsrörelsen använder sociala medier för att sprida sitt budskap. Idag pågår covid-19 pandemin och på kort tid har vaccin tagits fram och befolkningen erbjuds möjligheten att vaccinera sig. Det finns personer som är kritiska till vaccinet och de restriktioner som regering och riksdag infört för att minska smittspridningen av SARS-CoV-2. Studiens syfte är att få en förståelse för hur antivaccinationsrörelsens arbete tar sig uttryck i sociala medier och hur antivaccinationsrörelsen agerar för att påverka människor. Den metod som ligger till grund för arbetet är netnografi och syftet med en sådan studie är att studera sociala och kulturella sammanhang på internet. För att samla in empiri har inlägg på Twitter studerats genom en icke-deltagande observation. För att analysera inläggen har Anthony Giddens begrepp tillit och expertsystem samt Ulrick Becks begrepp risksamhället använts. Studiens resultat visar på att individer som uttrycker sig negativt om vaccinet besitter ett lågt förtroende för de experter som uttalar sig gällande SARS-CoV-2 och vaccinet som utvecklats. De som är kritiska till vaccin menar på att vaccinet mot SARS-CoV-2 och de restriktioner som finns på grund av covid-19 pandemin är ytterligare ett sätt att begränsa människor fri- och rättigheter. De individer som är kritiska till vaccin menar vidare att covid-19 pandemin är skapad för att the big pharma ska tjäna pengar. Vaccinet mot covid-19 anses vara farligare än att genomgå sjukdomen och det är en anledning till att inte vaccinera sig. / The World Health Organization (WHO) has listed ten threats to the world health, two of which are pandemics and vaccine skepticism. Scientist agree that pandemics will affect humanity andat the same time people are being more skeptical about vaccine. Individuals who are critical of vaccine belongs to the anti-vaccination movement and believe that vaccines are a way of harming the population. The anti-vaccination movement use social media to spread their message. The ongoing pandemic is called covid-19 pandemic, vaccines have been developed in a short time and the population is offered the opportunity to be vaccinated. There are people who are critical of the vaccine and the restrictions that the government and the parliament hade decide about to reduce the spread of the SARS-CoV-2. The purpose of this study is to establish an understandig of how the work of the anti-vaccination movement is expressed on social media and how they act to influence people. The method for this study is netnography and it is used to study social and cultural contexts on the internet. To gather empirical data posts on Twitter have been studied through a non-participatory observation. Anthony Gidden’s concepts of trust and expert systems as well as Ulrich Beck’s concepts of the risk society have been used to analyze the posts. The results of this study show that individuals who express themselves negatively about vaccines have a low level of trust in the experts who are commenting about SARS-CoV-2 and they believe that vaccines are a way to control people. They believe that the restrictions that exist due to covid-19 pandemic are another way to restrict people’s freedom and that it was the big pharma that created the covid-19 pandemic. The anti-vaccination movement considered that the vaccine against the SARS-CoV-2 is more dangerous than the disease.

I Sverige förekommer diskussioner om valfusk på sociala medier. Vid valet 2018 låg Valmyndighetens hemsida nere efter en överbelastningsattack, och fyra år senare sprids fortfarande teorier om att Valmyndigheten själva stängt ner sin hemsida för att påverka valresultatet (SvD, 2022). Twitter är en plattform där diskussioner om valfusk sker och där svensk politik diskuteras, det råder en oerhörd polarisering på plattformen kring detta ämne. I denna uppsats har vi med hjälp av en explorativ kartläggning undersökt hur förtroendet till demokratin möjligtvis kan urholkas genom att titta på förekomsten av narrativ om valfusk på Twitter, vår forskningsfråga lyder "hur och i vilken omfattning pratar man om valfusk innan, under och efter det svenska valet 2022". Vi har använt oss av en kvalitativ innehållsanalys för att analysera data från Twitter som handlar om valfusk under tidsperioden 1/8-22 fram till och med den 9/10-22. Vi har delat upp datasetet i tre delar, en period innan valet, en period som sträcker sig under valveckan samt en period som är tiden efter valet fram till den 9/10-22. Vi har kombinerat den kvalitativa innehållsanalysen med en kvantitativ innehållsanalys. Där vi genom deskriptiv statistik visar på förekomsten av de olika kategorierna vi tagit fram i materialet för att på så sätt identifiera i vilken utsträckning olika narrativ om valfusk förekommer i diskussioner om svenska valet 2022 på Twitter. Vårt resultat visar att tweetsen kring valfusk skiljer sig åt beroende på vilken period under valet som undersökts. I perioden innan valet var tweets som relaterar till att valfusk sker 66,5 procent av av det analyserade materialet. Dessa siffror skiftade i perioden under valet där motstånd mot narrativ om valfusk växte och utgjorde nästan lika stor andel som de som ansåg att valfusk sker (31 respektive 40,6 procent). Efter valet identifierade vi en ny kategori av tweets som var den näst största under den perioden (30,5 procent), efter de som anser att valfusk sker i svenska val. Det var de som ansåg att tweets om valfusk hade ökat om oppositionen ej vunnit riksdagsvalet, och att det istället blivit väldigt tyst från de individer som tidigare varit högljudda om att valfusk sker i svenska val. Diskussionen om valfusk är polariserad, där det är två ytterligheter som tampas mot varandra. 50 procent av det totala antalet inlägg som analyserats handlar om att valfusk pågår eller förväntas ske i svenska val, bland dessa inlägg riktas misstro till den svenska demokratin och det svenska valsystemet. Det uttrycks ett stort missnöje och hat mot de s.k "uråldriga" statliga medieinstitutionerna som anklagas för att vara en propagandamaskin för vänsterblockets räkning. Socialdemokraterna är det parti som överlägset mest anklagas för valfusk i debatten, 94,5 procent av alla inlägg som är riktade mot ett parti är riktade mot Socialdemokraterna. De anklagas för långvarig korruption, valfusk i tidigare val, där argumentet om att Socialdemokraterna är det enda partiet som blivit dömda för valfusk i domstol är vanligt förekommande. Huruvida narrativet om valfusk förändras för de olika tidsperioderna så visar resultatet att det är en större andel inlägg som påstår att valfusk sker i Sverige innan valet jämfört med under och efter valet. I datasetet efter valet handlar istället mycket kring diskussionen om valfusk om hur valresultatet hade gått åt andra hållet så hade det varit ett gigantiskt rop om valfusk från främst Sverigedemokrater / In Sweden, there are discussions about election fraud. At the 2018 election, the Swedish Election Authority's website was down after an overload attack, and four years later, theories and discussions are still spreading that the Swedish Election Authority itself shut down its website in order to influence the election results (SvD, 2022). Twitter is a platform where discussions about election fraud take place and where Swedish politics is discussed, there is an enormous polarization on the platform around this topic. In this essay, with the help of an exploratory survey, we have investigated how trust in democracy can possibly be eroded by looking at the prevalence of the spread of narratives about election fraud on Twitter, our research question is "how and to what extent do you talk about election fraud before, during and after the Swedish election in 2022". We have used a qualitative content analysis to analyze data from Twitter dealing with election fraud during the time period 1/8-22 up to and including 9/10-22. We have divided the data set into three parts, a period before the election, a period that extends during the election week and a period that is the time after the election until 9/10-22. We have combined the content of the qualitative analysis with a quantitative content analysis. Where we use descriptive statistics to show the existence of the different categories we developed in the material in order to identify the extent to which different narratives about electoral fraud appear in discussions about the Swedish 2022 election on Twitter. Our results show that the tweets about electoral fraud differ depending on which period during the election was examined. In the period before the election, tweets relating to election fraud were 66.5 percent of the analyzed material. These numbers shifted in the period during the election where opposition to the narrative of electoral fraud grew and constituted almost as large a proportion as those who believed that electoral fraud was taking place (31 and 40.6 percent, respectively). After the election, we identified a new category of tweets that was the second largest during that period (30.5 percent), after those who believe that electoral fraud occurs in Swedish elections. They were the ones who believed that tweets about electoral fraud would have increased if the opposition had not won the parliamentary election, instead the individuals who had previously been vocal about electoral fraud went silent. The discussion about electoral fraud is polarized, where two extremes are pitted against each other. 50 percent of the total number of posts analyzed are about electoral fraud taking place or expected to take place in Swedish elections, among these posts, a huge amount of mistrust of Swedish democracy and the Swedish electoral system is identified. There is great dissatisfaction and hatred expressed towards the so-called "ancient" state media institutions which are accused of being a propaganda machine on behalf of the left side. The Social Democrats are by far the party most accused of electoral fraud in the debate, 94.5 percent of all posts directed at any party are directed at the Social Democrats. They are accused of long-term corruption, electoral fraud in previous elections, where the argument that the Social Democrats are the only party that has been convicted of electoral fraud in court is common. Whether the narrative about election fraud changes for the different time periods, the results show that there is a greater proportion of posts that claim that election fraud takes place in Sweden before the election compared to during and after the election. In the data set after the election, instead, much of the discussion about election fraud is about how the election results had gone the other way, so there would have been a gigantic outcry about election fraud from mainly the Sweden Democrats.

[pt] Desastres variam em diferentes aspectos, como intensidade, impactos, tipo, duração, início e consequências. É um grande desafio buscar uma metodologia de avaliação de custos econômicos de desastres, sejam de pequenas ou grandes proporções, de início súbito ou lento, bem como para gerenciar seus respectivos custos (diretos, indiretos, interrupção de serviços, intangíveis). Dentro deste contexto, as mídias sociais, por serem uma fonte de dados com informações públicas, disponíveis em tempo real e com baixo custo de coleta, tornam-se fontes de dados importantes neste processo. Desta forma, o objetivo principal desta tese é propor uma metodologia de avaliação econômica de desastres baseada nas informações coletadas através das mídias sociais. Para atingir este objetivo, revisões sistemáticas da literatura (RSL) foram realizadas para compreender o estado da arte das metodologias de avaliação existentes e analisar como as mídias sociais podem ser utilizadas para suportar a avaliação econômica de desastres. Os resultados das RSL fornecem uma taxonomia, um framework e uma agenda de pesquisa que servem de base para criação de uma inédita metodologia de avaliação de custos econômicos de desastres, denominada AVECODE. Composta por três fases (Pré-evento, Desastre, Pós-evento) e duas áreas transversais (Coordenação e Tecnologia), a AVECODE é aplicada em um estudo de caso ocorrido no Rio de Janeiro em 2019, através do desenvolvimento de um protótipo e de dados coletados do Twitter. Os resultados desta aplicação são, então, comparados e analisados com a avaliação realizada pelo governo local. A AVECODE mostra-se uma importante ferramenta para concepção e implementação de políticas públicas, onde os resultados obtidos permitem a rastreabilidade das informações coletadas, a identificação de custos não contabilizados e averiguações nos quantitativos e custos gerados pelo governo local. Finalmente, a aplicação da metodologia em um estudo de caso permite ratificar a correlação da metodologia proposta com a prática. / [en] Disasters vary in different aspects, such as intensity, impacts, type, duration, onset, and consequences. Therefore, it is a great challenge to seek an economical cost assessment methodology to deal with disasters, whether small or large, of sudden or slow onset, as well as to manage their costs (direct, indirect, service interruption, intangibles). Within this context, social media, as a source of data with public information, available in real-time and with a low cost of collection, become a crucial data source in this process. Thus, the main objective of this thesis is to propose a methodology for the economic assessment of disasters based on information collected through social media. For that purpose, a systematic literature review (SLR) was carried out to understand the state of the art of existing assessment methodologies and to analyze how social media can be applied to support economic disaster assessment. The results of the SLR provide a taxonomy, a framework, and a research agenda that serve as a basis for creating an original methodology for assessing economic costs of disasters, named AVECODE. Comprised of three phases (Pre-event, Disaster, Post-event) and two transversal areas (Coordination and Technology), AVECODE is applied in a case study that took place in Rio de Janeiro in 2019, through the development of a prototype and data collected from Twitter. The results of this application are compared and analyzed with the disaster evaluation carried out by the local government. AVECODE proves to be a valuable tool for the design and implementation of public policies, where the results obtained allow the traceability of the collected information, the identification of unaccounted costs, and inquiries in the amounts and costs estimated by the local government. Finally, the application of the methodology in a case study allows us to confirm the correlation between the proposed methodology and practice.

This study analyzes peoples reactions in social media to the release of a company’s quarterly report. Sentiment analysis was performed on tweets about a company both from a short-and long-term perspective. On the long-term perspective, a two year period of sentiment was studied in relation to its quarter’s percental change of net income and revenues. Three companies was investigated in this process. On the short-term perspective, a linear regression was conducted based the quarter’s percental change of net income and revenues and on sentiment score, ranging from release day to 2 days after the release of a company’s quarterly report, in total five companies were studied. The result inferred that there was no correlation between the company’s net income and the reaction on Twitter on both long and short-term. Our conclusion is that the attitude towards a company is influenced by other factors than net income and revenues. The study also indicates that sentiments found in Twitter messages regarding a company name are related to a positive or negative expectation of the company. Findings suggest future studies to focus on companies greatly dependent on one product and analyze sentiment regarding that product instead, as the product is likely to impact financial results. / Denna studie undersöker hur människors reaktioner utspelas på sociala medier efter att ett företags kvartalsrapport släppts. En sentimentanalys utfördes på tweets angående ett företag ur både ett kortsiktigt och ett långsiktigt perspektiv. I det långsiktiga perspektivet jämfördes sentimenten från en tvåårsperiod med resultatet från kvartalsrapporterna under samma tid. Tre företag undersöktes. I det kortsiktiga perspektivet utfördes en linjär regression baserat på förändringen av rörelseresultatet och intäkterna, samt sentimentet med ett tidsintervall på 2 dagar från det att bolagets kvartalsrapport publiceras, detta utfördes på fem bolag. Resultaten från både det långsiktiga och det kortsiktiga perspektivet tyder på att det inte finns någon korrelation mellan företagets rörelseresultat och intäkter, samt sentimentet på Twittermeddelanden innehållande bolagets namn. Vår slutsats är att inställningen till ett företag påverkas av andra faktorer än rörelseresultatet och intäkterna. Studien indikerar också att inställningen på Twitter till ett företagsnamn är baserat på en positiv eller negativ förväntan. Vi föreslår att framtida studier fokuserar på företag som är starkt beroende av en produkt och analyserar sentiment för den specifika produkten istället eftersom produkten i detta läge har en stor inverkan på kvartalsrapportens resultat.

Our study is a collaborative dissertation paper that combines two differentdiscourse analyses, textual and visual, based on a common theoretical background.The introduction guides the reader through the content of the study, at the same timeoffering a brief context of research. The aim of the paper is to address a gap that weidentified in the study of mindfulness, namely a critical approach, from a media andcommunication perspective, of how this concept is represented in social media. Eventhough our research questions are developed separately in the analyses conductedindependently, they can be reduced to three core questions: ‘How is the meaning ofmindfulness constructed on Twitter?’, ‘Are there any power relations in theconstruction of discourse and if they exist, how do they shape the discourse?’, ‘Howdoes the reproduction and circulation of discourse shape its meaning throughintertextuality?’For answering these questions existing research from psychology,sociology and business has been reviewed, with the mention that no relevantmedia and/or communication studies on mindfulness have been found.Therefore, our attempt to open a discussion in the field required a theoreticalframe of analysis. For that we chose Michel Foucault’s discourse theory, addingobservations on relations of power, and Stuart Hall’s theories of representation.The methodologies used for the two analyses are Fairclough’s and Rose’sapproaches of applied discourse analysis. Critical Discourse Analysis (CDA) andVisual Discourse Analysis (VDA) are two detailed disseminations of qualitativedata, conducted separately. Results show that there is a mainstream discoursethat portrays mindfulness as a positive practice. This type of discourse might beinvested with power, however our conclusions in this sense are restrained by thelimitations of access to Twitter data. High intertextuality and low reliability onthe scientific discourse further suggested in our case that the understandings ofmindfulness are subject to change due to an advanced grade of interpretabilityamong Twitter users.

The opinion of customers about companies are very important as this can influence a company’s profit. Companies often get customer feedback via surveys or other official methods in order to improve their services. However, some customers feel threatened when their opinions are publicly asked and thus prefer to voice their opinion on the internet where they take comfort in anonymity. This form of customer feedback is difficult to monitor as the information can be found anywhere on the internet and new information is generated at an astonishing rate. Currently there are companies such as Brandseye and Brand.Com that provide online reputation management services. These services have various shortcomings such as cost and is incapable of accessing historical data. Companies are also not allowed to purchase these software and can only use the software on a subscription basis. The design proposed in this document will be able to scan any number of user defined websites and save all the information found on the websites in a series of index files, which can be queried for occurrences of user defined keywords at any time. Additionally, the software will also be able to scan Twitter and Facebook for any number of user defined keywords and save any occurrences of the keywords to a database. After scanning the internet, the results will be passed through a similarity filter, which will filter out insignificant results as well as any duplicates that might be present. Once passed through the filter the remaining results will be analysed by a sentiment analysis tool which will determine whether the sentence in which the keyword occurs is positive or negative. The analysed results will determine the overall reputation of the keyword that was used. The proposed design has several advantages over current systems: - By using the modular design several tasks can execute at the same time without influencingeach other. For example; information can be extracted from the internet while existing resultsare being analysed. - By providing the keywords and websites that the system will use the user will have full controlover the online reputation management process. - By saving all the information contained in a website the user will be able to take historicalinformation into account to determine how the keywords reputation changes over time. Savingthe information will also allow the user to search for any keyword without rescanning theinternet. The proposed system was tested and successfully used to determine the online reputation of many user defined keywords. / MIng (Computer and Electronic Engineering), North-West University, Potchefstroom Campus, 2015

The opinion of customers about companies are very important as this can influence a company’s profit. Companies often get customer feedback via surveys or other official methods in order to improve their services. However, some customers feel threatened when their opinions are publicly asked and thus prefer to voice their opinion on the internet where they take comfort in anonymity. This form of customer feedback is difficult to monitor as the information can be found anywhere on the internet and new information is generated at an astonishing rate. Currently there are companies such as Brandseye and Brand.Com that provide online reputation management services. These services have various shortcomings such as cost and is incapable of accessing historical data. Companies are also not allowed to purchase these software and can only use the software on a subscription basis. The design proposed in this document will be able to scan any number of user defined websites and save all the information found on the websites in a series of index files, which can be queried for occurrences of user defined keywords at any time. Additionally, the software will also be able to scan Twitter and Facebook for any number of user defined keywords and save any occurrences of the keywords to a database. After scanning the internet, the results will be passed through a similarity filter, which will filter out insignificant results as well as any duplicates that might be present. Once passed through the filter the remaining results will be analysed by a sentiment analysis tool which will determine whether the sentence in which the keyword occurs is positive or negative. The analysed results will determine the overall reputation of the keyword that was used. The proposed design has several advantages over current systems: - By using the modular design several tasks can execute at the same time without influencingeach other. For example; information can be extracted from the internet while existing resultsare being analysed. - By providing the keywords and websites that the system will use the user will have full controlover the online reputation management process. - By saving all the information contained in a website the user will be able to take historicalinformation into account to determine how the keywords reputation changes over time. Savingthe information will also allow the user to search for any keyword without rescanning theinternet. The proposed system was tested and successfully used to determine the online reputation of many user defined keywords. / MIng (Computer and Electronic Engineering), North-West University, Potchefstroom Campus, 2015