abstract: Learning analytics application is evolving into a student-facing solution. Student-facing learning analytics dashboards (SFLADs), as one popular application, occupies a pivotal position in online learning. However, the application of SFLADs faces challenges due to teacher-centered and researcher-centered approaches. The majority of SFLADs report student learning data to teachers, administrators, and researchers without direct student involvement in the design of SFLADs. The primary design criteria of SFLADs is developing interactive and user-friendly interfaces or sophisticated algorithms that analyze the collected data about students’ learning activities in various online environments. However, if students are not using these tools, then analytics about students are not useful. In response to this challenge, this study focuses on investigating student perceptions regarding the design of SFLADs aimed at providing ownership over learning. The study adopts an approach to design-based research (DBR; Barab, 2014) called the Integrative Learning Design Framework (ILDF; Bannan-Ritland, 2003). The theoretical conjectures and the definition of student ownership are both framed by Self-determination theory (SDT), including four concepts of academic motivation. There are two parts of the design in this study, including prototypes design and intervention design. They are guided by a general theory-based inference which is student ownership will improve student perceptions of learning in an autonomy-supportive SFLAD context. A semi-structured interview is used to gather student perceptions regarding the design of SFLADs aimed at providing ownership over learning. / Dissertation/Thesis / Masters Thesis Educational Psychology 2019

Der vorliegende Auswertungsbericht fasst die Ergebnisse einer Studie der TU Dresden zum Einfluss von Analytics Tools auf das Controlling der 3.000 größten Unternehmen in Deutschland im Jahr 2021 zusammen. Der Auswertungsbericht gibt einen Überblick über den Stand der Gestaltung und der Nutzung von Analytics Tools im Controlling. Befragt wurden die in den Unternehmen verantwortlichen Controllingleiter bzw. kaufmännische Geschäftsführer und CFOs mittels eines strukturierten Fragebogens. Der Rücklauf von 322 verwertbaren Fragebögen bei einer Rücklaufquote von 10,78 % unterstreicht das große Interesse der Praxis an dem Untersuchungsthema.:Inhaltsverzeichnis Abbildungsverzeichnis 1 Einleitung 1.1 Zielsetzung und untersuchte Aspekte 1.2 Inhalte des Auswertungsberichts und weitere Schritte im Forschungsprojekt 2 Grundkonzepte der Studie: Ein theoretischer Überblick 2.1 Der Begriff der Digitalisierung 2.1.1 Big Data als Grundlage für Business Analytics 2.1.2 Business Analytics 2.1.3 Abgrenzung von Business Analytics zu anderen Technologien 2.1.4 Business Analytics im Controlling 2.2 Psychologische Effekte von Digitalisierung (Rollenstress) 3 Datenerhebung und Auswertungsmethodik 3.1 Charakterisierung der Grundgesamtheit 3.2 Ablauf der Datenerhebung 3.3 Zusammenfassung des Fragebogenrücklaufs 3.4 Auswertungsmethodik 4 Empirische Ergebnisse zur Nutzung und Gestaltung von Analytics Tools im Controlling 4.1 Demografie der Antwortenden 4.2 Teil 1: Generelle Fragen zum Unternehmen 4.2.1 Organisatorische Einbettung des Controllings 4.2.2 Stand der Digitalisierung des Controllings im Unternehmen 4.2.3 Beitrag der Controlling-Abteilung für das Unternehmen 4.2.4 Einfluss der Corona-Pandemie 4.2.5 Veränderungen im Unternehmensumfeld 4.3 Teil 2: Fragen zur Controlling-Abteilung und zum Einsatz von Analytics Tools im Controlling 4.3.1 Aktivitäten der Controlling-Mitarbeiter (Rollenverständnis) 4.3.2 Verwendete Analytics Tools 4.3.3 Effekte der Analytics Tools 4.3.4 Art der Nutzung von Analytics Tools 4.3.5 Ressourcen für Analytics-Initiativen 4.3.6 Datenorientierung und Datenkultur 4.3.7 Big Data-Charakteristik der Daten 4.3.8 Eigenschaften von in Analytics Tools genutzten Daten 4.3.9 Technologische Charakteristika der Analytics Tools 4.3.10 Unterstützung durch das Top Management Team 4.3.11 Fähigkeiten der Führungskräfte im Controlling 4.3.12 Technische Fähigkeiten von Controlling-Mitarbeitern 4.3.13 Analytische Fähigkeiten der Controlling-Mitarbeiter 4.3.14 Wissenszugang und -nutzung 4.4 Teil 3: Fragen zum Einfluss von Analytics Tools auf die Tätigkeit und das Arbeitsumfeld von Controllingleitern 4.4.1 Auswirkungen von Informationen aus Analytics Tools 4.4.2 Arbeitsrelevante Informationen für die Tätigkeit als Controlling-Leiter 4.4.3 Umstände der Tätigkeit von Controllingleitern (Rollenüberlastung) 4.4.4 Wahrnehmungen der Arbeit von Controlingleitern (Rollenambiguität und Rollenkonflikt) 4.4.5 Einstellungen zum Unternehmen 4.5 Sonstige Hinweise der Studienteilnehmer 5 Management Summary 6 Literaturverzeichnis

Affärsanalys är en ökande trend som många organisationer idag använder på grund av potentialen att fastställa värdefulla insikter, ökad lönsamhet och förbättrad operativ effektivitet. Något som visat sig vara problematiskt då det önskade resultatet inte alltid är en självklarhet. Syftet med studien är att undersöka hur modeföretag kan använda datadriven affärsanalys för att generera positiva insikter genom värdeskapande mekanismer. Utifrån semistrukturerade intervjuer med anställda på ett modeföretag har vi, med utgångspunkt i tidigare forskning, kartlagt hur datadriven affärsanalys brukas för att skapa värde genom att applicera en processmodell på verksamheten. Empirin resulterade i tre värdefulla insikter (1) Det studerade företaget använder affärsanalys för ökad lönsamhet (2) Företagets data tillgångar är tillräckliga för att utvinna värdefulla insikter (3) Vidare såg vi att företaget arbetar med influencers vilket är en ny affärsanalys-funktion som inte definierats i tidigare forskning. / Business analysis is an increasingly popular trend that many organisations use because of its potential to establish valuable insights, increased profitability and improved operational efficiency. Something that has proved to be rather problematic as the desired results rarely is a certainty. The purpose of the study is to examine how fashion retailers can use business analytics to generate positive insights through value-creating mechanisms by applying a process model. Based on semi-structured interviews with the employees of a fashion company and a starting point in previous research, we have mapped how business analysis can be used to obtain value. The empirical study resulted in three valuable insights (1) The examined organisation uses business analysis to increase profitability. (2) The data assets of the organisation are enough to acquire valuable insights. (3) Further we discovered that the organisation uses influencers as a valuable asset and can be categorised as a business analysis capability, previously undefined in preceding research.

With globally increased migration and mobility between countries, it has become critical for many people to learn to speak a second language. The focus of this study is on adult migrant language learners that are learning a second language of the host country on the side of their working life. This study aims to support learners in their second language acquisition outside classrooms settings. In particular, it explores how the use of a specially designed application aimed at helping learners to keep track on how much time they spend on studying a second language affects their engagement and motivation to continue study the target language. To support migrant learners keeping track of the time spent on language learning activities (e.g., speaking, writing, reading and listening), a web-based application, the TimeTracker App, accessible through users’ mobile device has been developed by the researcher and offered to the learners. Participants in this study used the application for around two weeks. A mixed method approach was employed: data was collected through semi-structured interviews and by extracting log data from the application’s database. Interview data was analysed by means of a conventional content analysis and log data by using descriptive statistics. Overall, the study’s results show that the use of the TimeTracker App enabled the respondents to feel more aware of how much time they spent on their studies, and inspired them to devote more time to study the target language compared to before using the application. The findings suggest that migrant learners become more motivated and engaged in their second language learning when using the application. / Globalt ökad migration och rörlighet mellan länder har gjort det kritiskt för många att lära sig att tala ett andraspråk. Denna studie fokuserar på arbetande migranter som lär sig ett andraspråk vid sidan av sitt arbetsliv. Studien syftar till att stödja de studerande i sitt lärande av ett andraspråk utanför klassrummet. I synnerhet undersöker den hur användningen av en speciellt utformad applikation som syftar till att hjälpa eleverna att hålla reda på hur mycket tid de spenderar på att studera ett andraspråk påverkar deras engagemang och motivation för att fortsätta studera målspråket. För att hjälpa studerande migranter hålla reda på den tid som spenderas på språkinlärning (t ex att tala, skriva, läsa och lyssna) har en webbaserad applikation, TimeTracker App, som är tillgänglig via användarnas mobiltelefon, utvecklats av författaren och erbjudits till eleverna. Deltagarna i denna studie använde applikationen i cirka två veckor. En blandad metod användes: data samlades in genom halvstrukturerade intervjuer och genom att extrahera loggdata från applikationsdatabasen. Intervjudata analyserades med hjälp av en konventionell innehållsanalys och loggdata med hjälp av beskrivande statistik. Sammantaget visar studiens resultat att användningen av TimeTracker App gjorde det möjligt för respondenterna att bli mer medvetna om hur mycket tid de spenderade på sina studier och det inspirerade dem att ägna mer tid att studera målspråket jämfört med innan man använde applikationen. Resultaten tyder på att arbetande migranter blir mer motiverade och engagerade i sitt studerande av ett andraspråk när de använder applikationen.

[pt] Cursos online têm ampliado as possibilidades de pesquisa sobre comportamento e performance de estudantes. Esta tese investiga como apoiar instrutores na análise de logs de estudantes em Ambientes Virtuais de Aprendizagem. Primeiro, conduzimos entrevistas com instrutores e realizamos um mapeamento sistemático do estado da arte sobre Education Data Mining e Learning Analytics. Em seguida, analisamos logs de cursos online oferecidos no Brasil e comparamos nossas descobertas com resultados apresentados na literatura. Além disso, capturamos as preferências dos instrutores em relação a visualização de comportamento e performance de estudantes. Contudo, notamos uma lacuna de trabalhos mostrando modelos para o desenvolvimento de ferramentas de Learning Analytics. Com base nesses estudos, esta tese apresenta um modelo conectando teorias e modelos de visualização, assim como requisitos dos instrutores, suas preferências de visualização, diretrizes da literatura e métodos para análise de logs dos estudantes. Instanciamos e avaliamos esse modelo em uma ferramenta para montar dashboards, capturamos evidências de aceitação da nossa proposta e obtivemos feedbacks dos instrutores sobre a ferramenta tais como suas preferências de análise e visualizações. Por fim, apresentamos algumas considerações e discutimos lacunas existentes no trabalho que podem fundamentar e guiar futuras pesquisas, tais como desenvolvimento de novas instâncias e implantações do nosso modelo em instituições de ensino brasileiras e avaliação de eventuais mudanças na performance dos estudantes quando instrutores visualizam informações sobre o comportamento e performance deles, e agem de acordo. É importante ressaltar que a maioria dos estudos apresentados nessa tese foram conduzidos antes da pandemia de COVID-19. Somente o último estudo foi executado no início da pandemia no Brasil. / [en] Online education has broadened the avenues of research on student s behavior and performance. In this thesis, we shed light on how to support instructors in analyzing student logs from Virtual Learning Environments. Firstly, we conducted interviews with instructors and a systematic mapping of the state-of-art about Education Data Mining and Learning Analytics. Then, we analyzed logs from online courses offered in Brazil and compared our findings with results presented in the literature. Moreover, we gathered instructors preferences in regard to visualization of both students behavior and performance. However, we noted a lack of work showing models to support the development of learning analytics tools. In order to bridge this gap, this thesis presents a model connecting both Visual Analytics theories and models as well as instructors requirements, their visualization preferences, literature guidelines and methods for analyzing student logs. We instantiated and evaluated this model in a tool to assemble dashboards. We captured evidence of their acceptance of our proposal and obtained instructors feedback about the tool such as their both analysis and visualization preferences. Finally, we present some considerations and discuss gaps in existing works that can ground and guide future research, such as new instances of our model, as well as deploying them at Brazilian institutions and evaluating whether there are changes in students performance when instructors are able to see information about their behavior and performance, and act accordingly. It is worth highlighting that the majority of studies presented in this thesis were conducted before the COVID-19 pandemic. Only the last study was performed in the beginning of the pandemic in Brazil.

Projektet genomfördes i samarbete med Siemens Healthineers i syfte att utreda möjligheter till att prognostisera vårdflöden. Det genom att undersöka hur big data tillsammans med maskininlärning kan utnyttjas för prediktiv analys. Projektet utgjordes av två fallstudier med mål att, baserat på data från tidigare MRT-undersökningar, förutspå undersökningstider för kommande undersökningar respektive identifiera patienter som riskerar att missa inbokad undersökning. Fallstudierna utfördes med hjälp av programmeringsspråket R och tre olika inbyggda funktioner för maskininlärning användes för att ta fram prediktiva modeller för respektive fallstudie. Resultaten från fallstudierna gav en indikation på att det med en större datamängd av bättre kvalitet skulle vara möjligt att förutspå undersökningstider och vilka patienter som riskerar att missa sin inbokade undersökning. Det talar för att den här typen av prediktiva analyser kan användas för att prognostisera vårdflöden, något som skulle kunna bidra till ökad effektivitet och kortare väntetider i vården. / This project was performed in cooperation with Siemens Healthineers. The project aimed to investigate possibilities to forecast healthcare processes by investigating how big data and machine learning can be used for predictive analytics. The project consisted of two separate case studies. Based on data from previous MRI examinations the aim was to investigate if it is possible to predict duration of MRI examinations and identify potential no show patients. The case studies were performed with the programming language R and three machine learning methods were used to develop predictive models for each case study. The results from the case studies indicate that with a greater amount of data of better quality it would be possible to predict duration of MRI examinations and potential no show patients. The conclusion is that these types of predictive models can be used to forecast healthcare processes. This could contribute to increased effectivity and reduced waiting time in healthcare.

Sport clubs are always searching for innovative ways to improve performance and obtain a competitive edge. Sports analytics today is focused primarily on evaluating metrics thought to be directly tied to performance. Injuries indirectly decrease performance and cost substantially in terms of wasted salaries. Existing sports injury research mainly focuses on correlating one specific feature at a time to the risk of injury. This paper provides a multidimensional approach to non-contact injury prediction in Swedish professional ice hockey by applying machine learning on historical data. Several features are correlated simultaneously to injury probability. The project’s aim is to create an injury predicting algorithm which ranks the different features based on how they affect the risk of injury. The paper also discusses the business potential and strategy of a start-up aiming to provide a solution for predicting injury risk through statistical analysis. / Idrottsklubbar letar ständigt efter innovativa sätt att förbättra prestation och erhålla konkurrensfördelar. Idag fokuserar data- analys inom idrott främst på att utvärdera mätvärden som tros vara direkt korrelerade med prestation. Skador sänker indirekt prestationen och kostar markant i bortslösade spelarlöner. Tidigare studier på skador inom idrotten fokuserar huvudsakligen på att korrelera ett mätvärde till en skada i taget. Den här rapporten ger ett multidimensionellt angreppssätt till att förutse skador inom svensk elitishockey genom att applicera maskininlärning på historisk data. Flera attribut korreleras samtidigt för att få fram en skadesannolikhet. Målet med den här rapporten är att skapa en algoritm för att förutse skador och även ranka olika attribut baserat på hur de påverkar skaderisken. I rapporten diskuteras även affärsmöjligheterna för en sådan lösning och hur en potentiell start-up ska positionera sig på marknaden.

[pt] Game Analytics é uma área que envolve o processamento de dados de videogames com a finalidade de proporcionar uma melhor experiência de jogo para o usuário. Também ajuda a verificar os padrões de comportamento dos jogadores, facilitando a identificação do público-alvo. A coleta de dados dos jogadores ajuda os desenvolvedores de jogos a identificar problemas mais cedo e saber por que os jogadores deixaram o jogo ou continuaram jogando. O comportamento desses jogadores geralmente segue um padrão, fazendo com que se encaixem em diferentes perfis de jogadores. Especialistas em análise de jogos criam e usam modelos de tipos de jogadores, geralmente variantes do modelo de Bartle, para ajudar a identificar perfis de jogadores. Esses especialistas usam algoritmos de agrupamento para separar os jogadores em grupos diferentes e identificáveis, rotulando cada grupo com o tipo de perfil definido pelo modelo proposto. O objetivo principal deste projeto é criar um plugin Unity genérico para ajudar a identificar perfis de jogadores em jogos. Este plugin usa uma API Python, que lida com os dados do jogo armazenados em um banco de dados MongoDB, para agrupar e rotular cada partida ou nível do jogo escolhido enquanto o jogo está em execução. Neste plugin, os desenvolvedores de jogos podem configurar o número de tipos de jogadores que desejam identificar, os rótulos dos jogadores e até os algoritmos que desejam usar. Essa abordagem de agrupamento online não é usual no desenvolvimento de jogos. Até onde sabemos, não há nenhum componente de software na literatura de análise de jogos com a mesma direção e recursos. / [en] Game Analytics is an area that involves the processing of video game data, in order to make a better game experience for the user. It also helps to check the patterns in players behaviour, making it easier to identify the target audience. Gathering player data helps game developers identify problems earlier and know why players left the game or kept playing. These players behavior usually follows a pattern, making them fit in different player profiles. Game analytics experts create and use models of player types, usually variants of Bartle s model, to help identify player profiles. These experts use clustering algorithms to separate players into different and identifiable groups, labeling each group with the profile type defined by the proposed model. The main goal of this project is to create a generic Unity plugin to help identify Player Profiles in games. This plugin uses a Python API, which deals with the game data stored in a MongoDB database, to cluster and label each match or level of the chosen game while the game is running. In this plugin, game developers can configure the number of player types they want to identify, the player labels, and even the algorithms they wish to use. This online clustering approach is not usual in game development. As far as we are aware, there is no software component in the game analytics literature with the same direction and features.

Cultural data can come in various forms and modalities, such as text traditions, artworks, music, crafted objects, or even as intangible heritage such as biographies of people, performing arts, cultural customs and rites. The assignment of metadata to such cultural heritage objects is an important task that people working in galleries, libraries, archives, and museums (GLAM) do on a daily basis. These rich metadata collections are used to categorize, structure, and study collections, but can also be used to apply computational methods. Such computational methods are in the focus of Computational and Digital Humanities projects and research. For the longest time, the digital humanities community has focused on textual corpora, including text mining, and other natural language processing techniques. Although some disciplines of the humanities, such as art history and archaeology have a long history of using visualizations. In recent years, the digital humanities community has started to shift the focus to include other modalities, such as audio-visual data. In turn, methods in machine learning and computer vision have been proposed for the specificities of such corpora. Over the last decade, the visualization community has engaged in several collaborations with the digital humanities, often with a focus on exploratory or comparative analysis of the data at hand. This includes both methods and systems that support classical Close Reading of the material and Distant Reading methods that give an overview of larger collections, as well as methods in between, such as Meso Reading. Furthermore, a wider application of machine learning methods can be observed on cultural heritage collections. But they are rarely applied together with visualizations to allow for further perspectives on the collections in a visual analytics or human-in-the-loop setting. Visual analytics can help in the decision-making process by guiding domain experts through the collection of interest. However, state-of-the-art supervised machine learning methods are often not applicable to the collection of interest due to missing ground truth. One form of ground truth are class labels, e.g., of entities depicted in an image collection, assigned to the individual images. Labeling all objects in a collection is an arduous task when performed manually, because cultural heritage collections contain a wide variety of different objects with plenty of details. A problem that arises with these collections curated in different institutions is that not always a specific standard is followed, so the vocabulary used can drift apart from another, making it difficult to combine the data from these institutions for large-scale analysis. This thesis presents a series of projects that combine machine learning methods with interactive visualizations for the exploratory analysis and labeling of cultural data. First, we define cultural data with regard to heritage and contemporary data, then we look at the state-of-the-art of existing visualization, computer vision, and visual analytics methods and projects focusing on cultural data collections. After this, we present the problems addressed in this thesis and their solutions, starting with a series of visualizations to explore different facets of rap lyrics and rap artists with a focus on text reuse. Next, we engage in a more complex case of text reuse, the collation of medieval vernacular text editions. For this, a human-in-the-loop process is presented that applies word embeddings and interactive visualizations to perform textual alignments on under-resourced languages supported by labeling of the relations between lines and the relations between words. We then switch the focus from textual data to another modality of cultural data by presenting a Virtual Museum that combines interactive visualizations and computer vision in order to explore a collection of artworks. With the lessons learned from the previous projects, we engage in the labeling and analysis of medieval illuminated manuscripts and so combine some of the machine learning methods and visualizations that were used for textual data with computer vision methods. Finally, we give reflections on the interdisciplinary projects and the lessons learned, before we discuss existing challenges when working with cultural heritage data from the computer science perspective to outline potential research directions for machine learning and visual analytics of cultural heritage data.

De Förenta Nationernas Agenda 2030 fastställer som ett delmål att säkerställa utbildning av hög kvalitet och främja livslångt lärande för alla som en del av arbetet för ett mer hållbart samhälle. Vikten av detta delmål blir särskilt tydlig i och med det observerbara sambandet mellan en fullständig gymnasieexamen och allmän hälsa i Sverige; gymnasiestudenter som går ut med en gymnasieexamen tenderar att erhålla bättre allmän hälsa. Learning Analytics är ett relativt nytt område inom utbildningsvetenskaplig forskning som syftar till att förbättra utbildning med hjälp av elevdata. Detta arbete undersökte vilken möjlig påverkan och begränsningar som förekommer vid implementering av en multipel linjär regressionsmodell utvecklad för en matematikkurs i en gymnasieskola. Vid utvecklingen av denna modell fastställdes tre signifikanta indikatorer för att förutsäga elevernas slutbetyg; Diagnos resultat,resultat på nationella proven och frånvaro. Prognosmodellen har utvärderats statistiskt varpå den visade sig vara tillförlitlig i 90% av bedömningarna, vilket inte är tillräckligt säkert för att användas i verkliga bedömningstillfällen eftersom lärare kräver att resultaten är obestridliga. Genom en fokusgruppsintervju med lärare granskas dessa resultat och deltagarna uttrycker sitt intresse för prognosmodeller tillsammans med en reflektion över elevers potentiella negativa reaktioner på en ogynnsam prognos. Utvärdering av modellen visar att den i dagsläget har en rimlig förmåga att förutsäga elevers slutbetyg men att det finns ett starkt behov av insamling av mer nyanserade data för att öka möjligheten till innovation i framtida arbeten / The 2030 Agenda establishes the goal to ensure quality education and promote lifelong learning opportunities for all. The importance of this goal becomes particularly clear when taking into account the link between upper secondary school graduation and general health in Sweden; Upper secondary school graduates tend to have better general health. Learning Analytics is a relatively new area of education research which aims to improve education using student data. This report examines the possible impact and limitations when implementing a multiple linear regression model developed for a mathematics course in an upper secondary school. In developing this model, three major indicators are established to be significant in predicting students' final grade; Diagnosis results, national test results and the amount of student absence. The model was statistically evaluated and found to be reliable in 90% of cases, which is not secure enough to be used in real assessment situations as teachers require the results to be indisputable. Through a focus group interview with teachers these results are evaluated and the participants establish their interest in predictive tools along with concerns for students' negative reactions to poor results. Evaluation of the model shows it has a reasonable ability to anticipate students' final grades but with a strong need for improvement in data collection methods and acquisition of more nuanced data to support greater possibility for innovation in future works.