Dissertations / Theses on the topic 'Maskininlärning'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Maskininlärning.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
Pettersson, Gustav, and John Almqvist. "Lavinprognoser och maskininlärning : Att prediktera lavinprognoser med maskininlärning och väderdata." Thesis, Uppsala universitet, Institutionen för informatik och media, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-387205.
Full textThis research project examines the feasibility of using machine learning to predict avalanche dangerby usingXGBoostand openly available weather data. Avalanche forecasts and meterological modelledweather data have been gathered for the six areas in Sweden where Naturvårdsverket throughlavin-prognoser.seissues avalanche forecasts. The avanlanche forecasts are collected fromlavinprognoser.seand the modelled weather data is collected from theMESANmodel, which is produced and providedby the Swedish Meteorological and Hydrological Institute. 40 machine learning models, in the form ofXGBoost, have been trained on this data set, with the goal of assessing the main aspects of an avalan-che forecast and the overall avalanche danger. The results show it is possible to predict the day to dayavalanche danger for the 2018/19 season inSödra Jämtlandsfjällenwith an accuracy of 71% and a MeanAverage Error of 0.256, by applying machine learning to the weather data for that region. The contribu-tion ofXGBoostin this context, is demonstrated by applying the simpler method ofLogistic Regressionon the data set and comparing the results. Thelogistic regressionperforms worse with an accuracy of56% and a Mean Average Error of 0.459. The contribution of this research is a proof of concept, showingfeasibility in predicting avalanche danger in Sweden, with the help of machine learning and weather data.
Ermakova, Natalia, and Alicia Bråtner. "Maskininlärning och kvantmekanik." Thesis, KTH, Skolan för teknikvetenskap (SCI), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-255825.
Full textThis thesis aims to use machine learning to solve for the ground state energy of the quantum system corresponding to the particle in a box. A radial basis function (RBF) network is used with Gaussian functions as the variational wave function. The weights in the network are updated so that the energy expectation value is minimized, which is carried out by using the variational Monte Carlo (VMC) method. The method using machine learning succeeds in finding the ground state energy for the particle in a box. The method also works when a perturbation in the form of a linear potential is added to the infinite potential well.
Axelsson, Filip, and Marcus Jirwe. "Maskininlärning för kvantmekaniska problem." Thesis, KTH, Skolan för teknikvetenskap (SCI), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-255698.
Full textIn this article we analyze whether machine learning can be used to help solve problemsin physics. This is examined by implementing an artificial neural network which is trainedto find the energy levels for the quantum harmonic oscillator with and without an externalelectric field. Radial basis functions were used to make the neural network. Monte Carlomethods were used for heavy calculations. The method was shown to work well in somecases but had problems for large electric fields. The problems that occured were that theconvergence became unstable, with leaps in the energy and that the system did not alwaysconverge to the right energy level.
L, Thunberg Christian, and Niklas Mannerskog. "Stochastic Gradient Descent inom Maskininlärning." Thesis, KTH, Skolan för teknikvetenskap (SCI), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-254799.
Full textSome tasks, like recognizing digits and spoken words, are simple for humans to complete yet hard to solve for computer programs. For instance the human intuition behind recognizing the number eight, ''\textit{8}'', is to identify two loops on top of each other and it turns out this is not easy to represent as an algorithm. With machine learning one can tackle the problem in a new, easier, way where the computer program learns to recognize patterns and make conclusions from them. In this bachelor thesis a digit recognizing program is implemented and the parameters of the stochastic gradient descent optimizing algorithm are analyzed based on how their effect on the computation speed and accuracy. These parameters being the learning rate $\Delta t$ and batch size $N$. The implemented digit recognizing program yielded an accuracy of around $95$ \% when tested and the time per iteration stayed constant during the training session and increased linearly with batch size. Low learning rates yielded a slower rate of convergence while larger ones yielded faster but more unstable convergence. Larger batch sizes also improved the convergence but at the cost of more computational power.
Vestberg, Albin. "Modellering av åsiktsdynamik med maskininlärning." Thesis, KTH, Skolan för teknikvetenskap (SCI), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-255690.
Full textOpinion dynamics is about how opinions of people, also called agents, in a group changes in time. The aim of this report was to find out if it is possible to use machine learning with artificial neural networks to, from a given initial condition, compute the opinion dynamics for the agents in the group. Thousands of random initial conditions were generated and the dynamics were calculated using a model from R. Hegselmann and U. Krause. The solutions were then used as data to train artificial neural networks. When the trained neural networks were used on new initial conditions it was found that these could, with varying results depending on the method and number of agents, recreate the opinion dynamics and give answers that were close to the real solutions. The number of agents used were between two and ten. The best results were obtained when groups with two agents were examined. When the number of agents increased the errors in the solutions with machine learning also increased, but the errors were still small enough so that the solutions can be considered as good recreations of the opinion dynamics.
Holmäng, Arvid, and Grothusen Axel von. "Intäktsestimering med hjälp av Maskininlärning." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-299340.
Full textThis work examines the possibility of estimating revenue for companies using machine learning. The data on which the models are based consists of points from the companies’ balance sheets and other public data. Since the research area is unexplored prior to this study, the main focus of this thesis is to explore which methods are most suitable for the task and which features in the dataset have the greatest impact on the models. In the study, the issue is investigated with the help of four different models; Random Forest regression, XGBoost, ordinary least squares method and Lasso. The models were evaluated with quantitative measures such as R2 score and mean absolute percentage error (MAPE). The algorithm and final model that performed best based on these measures were Random Forest regression with an average R2 score of 0,8197 and MAPE score of 0.3864. This study concludes that ensemble methods such as XGBoost and Random Forest are probably more suitable to use for this type of study compared to simpler regression models such as least squares method and Lasso. In conclusion, the models can contribute to the initial financial analysis of companies for which the income is unknown.
Dyremark, Johanna, and Caroline Mayer. "Bedömning av elevuppsatser genom maskininlärning." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-262041.
Full textToday, a large amount of a teacher’s workload is comprised of essay scoring and there is a large variability between teachers’ gradings. This report aims to examine what accuracy can be acceived with an automated essay scoring system for Swedish. Three following machine learning models for classification are trained and tested with 5-fold cross-validation on essays from Swedish national tests: Linear Discriminant Analysis, K-Nearest Neighbour and Random Forest. Essays are classified based on 31 language structure related attributes such as token-based length measures, similarity to texts with different formal levels and use of grammar. The results show a maximal quadratic weighted kappa value of 0.4829 and a grading identical to expert’s assessment in 57.53% of all tests. These results were achieved by a model based on Linear Discriminant Analysis and showed higher inter-rater reliability with expert grading than a local teacher. Despite an ongoing digitilization within the Swedish educational system, there are a number of obstacles preventing a complete automization of essay scoring such as users’ attitude, ethical issues and the current techniques difficulties in understanding semantics. Nevertheless, a partial integration of automatic essay scoring has potential to effectively identify essays suitable for double grading which can increase the consistency of large-scale tests to a low cost.
Jansson, Christofer, and Sebastian Karlsson. "Maskininlärning inom bokförings- och faktureringssystem." Thesis, Linnéuniversitetet, Institutionen för informatik (IK), 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:lnu:diva-66841.
Full textStudien har ett utformat syfte som handlar om att undersöka inom vilka ramar automatisering av processer inom bokförings- och faktureringssystem är möjligt med hjälp av maskininlärning, med anledning av att skapa en vägledning för när och var en implementation är relevant i systemen. För studien har en datainsamling utförts på ett företag som erbjuder bokförings- och faktureringssystem samt arbetar med maskininlärning för området. Hos verksamheten har intervjuer utförts för att fånga in relevant empiri som resultat. Aspekter som kategoriserats från insamlad data presenteras i möjligheter, implementering, etik och alternativa tillvägagångssätt. Tillsammans med litteraturstudier har kategorierna ur empirin analyserats och fyra ramar identifierats. Ramarna som identifierats är följande: konkurrenskraft, tekniska faktorer, kunskap och kompetens samt påverkan på yrkesgrupper.Varje ram innehåller information och exempel på faktorer som bör beaktas. Företag kan med hjälp av ramarna utreda viktiga aspekter gällande när och var en implementation av maskininlärning är möjlig och relevant i bokförings- och faktureringssystem.
Ogeborg, Marcus, and Vincent Widerberg. "Schemaläggning med hjälp av maskininlärning." Thesis, KTH, Data- och elektroteknik, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-208936.
Full textThis study has been analyzing if machine learning could be useful to work-relatedscheduling. The analysis was based on predictions generated by prototypes usingbusiness calendars. The business calendars were collected from two service and installationcompanies in the Stockholm region. An analysis was conducted regardingif the application could be practically applied to devices such as a smartphone. Theanalysis was based on tests regarding the prototypes required time to perform theirtasks.Three prototypes were developed with algorithms that made them predictive. Density-based Spatial Clustering of Applications with Noise (DBSCAN), Logistic Regressionand Weighted K-Nearest Neighbors (wKNN) were the implemented algorithms.DBSCAN was the best-performing algorithm according to the tests. However, a conclusioncould not be found concerning whether machine learning could be useful.The number of successful predictions did not exceed the number of available timeson concerned days, which was assumed as unsatisfying results. In addition, the prototypesneeded a significant amount of resources which could be a problem in practicaluse.
Fürsten, Cardell Joshua, and Maya Källström. "Maskininlärning inom digital marknadsföring : En studie om hur maskininlärning hjälper eller stjälper digitala marknadsföringsbyråer, och vilka faktorer som bör tas i beaktning vid användning av maskininlärning." Thesis, Södertörns högskola, Medieteknik, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:sh:diva-44948.
Full textProblem: Trots att AI-verktyg har kommit att underlätta samt påverka den digitala marknadsföringen på flertalet olika sätt, så finns det ett fortsatt behov för forskning kring hur AI kan underlätta marknadsföringsproblem samt hur användningen av AI-verktyg i marknadsföringssyfte kan underlättas för digitala marknadsföringsbyråer. Syfte: Studiens syfte är att påvisa hur och varför maskininlärning både hjälper och stjälper digitala marknadsföringsbyråer, samt vilka faktorer som digitala marknadsföringsbyråer behöver ta i beaktning om de avser att använda maskininlärning inom digital marknadsföring. Metod: Studien utgår från en kvalitativ forskningsstrategi, med ett induktivt tillvägagångssätt där sex genomförda semistrukturerade intervjuer utförts. Semistrukturerade intervjuer har genomförts för att skapa flexibilitet hos individen som blir intervjuad, och således kan nya insikter och tankar uppstå. Tematisk analys har använts som analysmetod för studien, där fokus legat på att identifiera mönster och teman. Slutsats: Maskininlärning hjälper digitala marknadsföringsbyråer inom digital marknadsföring med att effektivisera arbetsprocesser genom att bearbeta data och information, frigöra tid för anställda samt möjligheten att erhålla konkurrensfördelar. Maskininlärning stjälper digitala marknadsföringsbyråer inom digital marknadsföring genom att exkludera mänskliga deltagare inom arbetssystemet, minska förtroende för organisationen samt skapa obehag vid aktiv personalisering om det används oetiskt. Faktorer som bör tas i beaktning av digitala marknadsföringsbyråer om de avser att använda maskininlärning inom digital marknadsföring är att digitala marknadsföringsbyråer inte kan ersätta mänsklig interaktion, mänsklig integritet är viktigt i det mån att digitala marknadsföringsbyråer efterföljer lagar och regler, storleken av verksamheten avgör vilka prioriteringar som görs mellan att nyttja tidigare kunskap och ny teknologi, organisationer behöver vara transparenta och AI måste nyttjas genomgående i organisationen för att organisationen ska gynnas på lång sikt.
Dufberg, Johan. "Automatisk dokumentklassificering med hjälp av maskininlärning." Thesis, Örebro universitet, Institutionen för naturvetenskap och teknik, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:oru:diva-67228.
Full textTo manually handle and classify large quantities of text documents, takes a lot of time and demands a large staff, to use machine learning for this purpose is an alternative. This thesis aims to give the reader a fundamental insight in how automatic classification of texts work and give a quick overview of the most common algorithms used for this purpose. The examples that are shown uses news articles in English about tech and finance, but the thesis takes a start in the question about how mature the technique is for handling official Swedish documents. The first part is the scientific background on which the second part rests, here several algorithms and techniques are described which is used in practice later. The report does not aim to describe a product in any form but acts as a “proof of concept” for the use of text classification. Finally, the results from the tests are discussed, and one of the conclusions drawn is that when data is abundant a relatively simple classifier can perform close to equal to a technically more developed and complex classifier. If the performance of the classifier is related to the time taken this indicates that complex classifiers need hardware with high computational power and a fair bit of memory for the classifier to be viable.
Andersson, Henrik, Robin Andersson, Leif Eriksson, Alfred Hagberg, Jonathan Lundgren, Mustaf Musse, and Eric Nylander. "Maskininlärning för automatisk matchning av produkter." Thesis, Linköpings universitet, Institutionen för datavetenskap, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-148836.
Full textSandh, Oskar. "MASKININLÄRNING FÖR PREDIKTIV DATAANALYS OCH VISUALISERING." Thesis, Örebro universitet, Institutionen för naturvetenskap och teknik, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:oru:diva-79467.
Full textMaskininlärning ett stort inflytande över många branscher. Mycket forskning har gjorts inom detta ämne, och ännu mer kommer att göras i framtiden. Detta projekt syftar på att undersöka maskininlärningsmodeller i kontexten för ett detaljhandelsföretag. Analysen begränsas till tio olika regressionsvarianter som testas över 6 olika datamängder, indelade i ytterligare delmängder. Modellerna kommer att bedömas efter dess uträknade RMSE (Root Mean Square Error) värde och en jämförelse kommer att göras utifrån dessa. I slutändan så kommer vissa slutsatser att dras utifrån undersökningens resultat. Utöver denna analys, så kommer även rapporten att beskriva utvecklingen av en applikation som har syftet att presentera prediktionsinsikter för en kund. Stor fokus har inte legat på denna del under projektets gång, men rapporten kommer berätta om hur grunden och dess primära delar har utvecklats.
Carlstein, Mikael, and Kim Vigren. "Matcha platsannonser och arbetssökande med maskininlärning." Thesis, Malmö universitet, Fakulteten för teknik och samhälle (TS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:mau:diva-20920.
Full textThe paper aims to investigate the possibility of matching job seekers and job advertisements with general competencies. General competencies deal with how job seekers act in relation to others, solve their duties and act in relation to change. During the study, a pilot experiment and three experiments were conducted to investigate this. The results of the experiments provided an indication that it is possible but that further research and development is required to achieve better results. By counting the existence of words linked to general competencies in job advertisements, they can be classified according to which general competencies they seek, 12282 job listings were classified by this method. The match between job seekers and job advertisements received an accuracy of 48.57%, the low result indicates that further research on the comparison between a job applicant and job advertisement is recommended.
Järkeborn, Sandra, and Vera Werner. "Automatisering av kundtjänst med maskininlärning : Hur maskininlärning kan användas inom kundtjänst samt hur detta påverkar företagskultur och kundnöjdhet." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-241110.
Full textArtificial intelligence and machine learning hasbeen on the radar for a while now. What started as a groundbreaking phenomenon is now incorporated in people's everyday life. With all its benefits regarding personalization and “smart” features, there is also a concern for the world becoming too inhumane, impersonal and dependent of machines. This is an investigation touching upon these issues, applied on Parks & Resorts a highly valueand culture driven company. The purpose was to analyze the pros and cons of automating the customer support, and if it’s possible to do so without cutting corners, taking the company’s values into account. The results show that there are benefits of implementing a chat bot, especially when it comes to being effective, but also that implementing one that is good enough to attain current standards wouldrequire numerous resources.
Saveh, Diana. "Anpassning av mobilnotifikationer med hjälp av maskininlärning." Thesis, Mittuniversitetet, Institutionen för informationssystem och –teknologi, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:miun:diva-36451.
Full textDenna studie handlade om att försöka minska störande notifikationer som kan upplevas som stressande och irriterande. Det som skapades var en applikation som agerade som en notifikationskontroll. Denna applikation fungerar med hjälp av maskininlärning som ska förutse när användaren ville ta emot sina notifikationer. Den mönsterigenkännande artificiella intelligensen som användes kallas associationsregelanalys. Associationsregelanalysen använde sig av ett träd som kallas fp-growth. Det gjordes ett användartest före installation av applikationen och ett användartest efter för att se hur användaren upplevde stress men även själva applikationen. Studien visade att skärmtiden minskade med en timme och antalet gånger som mobilen öppnades minskades också. Denna undersökning kräver mer data då det kan vara så att användaren inte blev påverkad av applikationen utan endast slumpmässigt använde mobiltelefonen mindre.
Enerstrand, Simon. "Klassificering av kvitton med hjälp av maskininlärning." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-261144.
Full textMachine learning is used in more and more areas. It has the potential to replace many repetitive tasks, or at least simplify them. Document management within financial systems is an area machine learning can help with. A lot of manual input is often needed in different fields by reading invoices or receipts. The goal of the project is to create an application that uses machine learning for the company Centsoft AB. The application should receive OCR-interpreted texts from an image of a receipt and then, with high certainty, be able to determine which category the receipt belongs to. This report aims to show the development of the machine learning model in the application. The report answers the question: "How can receipts be classified using machine learning?".The methodology case study and the research method MoSCoW will be applied during the project. The project also considers the triangle method described by Eklund. Machine learning frameworks are used to evaluate the trained model. The trained model can, with high certainty, interpret receipts it has not encountered before. In order to get a meaningful interpretation, receipts must have the intention of belonging to one of the eight trained categories.The choice of methods suited the project well to answer the question. The application can be further developed and be implemented in the invoice management system. The implementation of the project gives knowledge about how to work with machine learning solutions. In the future, the technology can be applied in several areas.
Al-Wendawi, Kamal, and Dyar Faradj. "Smartare resursfördelning med hjälp av djup maskininlärning." Thesis, KTH, Hälsoinformatik och logistik, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-278066.
Full textAbstractThis thesis has been carried out on behalf of Stanley Security Sverige AB. The company Stanley provides a product to calculate the number of people in a queue. The product detects the number of people in a queue with the help of a surveillance camera and deep machine learning. They find that the existing solution does not perform well enough to be deployed and does not detect with high accuracy. In this work, various known object recognition models have been examined and compared, as well as examination of previous work and tests in the field.The result was a reconfigured model prototype that only detected faces in comparison to known models You Only Look Once V3 (YOLOV3) and Single Shot MultiBox Detector (SSD) that detect the entire human body, in addition to other objects. The proposed model performed significantly better than the known models and had higher accuracy. The results indicate that the proposed model prototype can be deployed in real time and be used by Stanely's customers to efficiently allocate resources within their companies.Keywordsmachine learning, image analysis, Python, Darknet, neural network, deep learning, YOLOV3, Single Shot MultiBox Detector.
Larsson, Therése, and Karl Paradis. "Litteraturstudie: Tillämpningen av maskininlärning vid algoritmisk handel." Thesis, Malmö universitet, Fakulteten för teknik och samhälle (TS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:mau:diva-20946.
Full textWe conduct a literature review in which we study and analyze publications in the area of machine learning in combination with algorithmic trading. In this study we investigate what types of data and which machine learning techniques that are shown to be applicable to systems used for algorithmic trading. For our literature review we use peer-reviewed publications from trustworthy databases. The result shows that we find mainly three types of data that are relevant for algorithmic trading. These are financial data quotes, technical indicators and the types of data that is relevant for fundamental analysis. Financial data quotes often seem to be used as a basis for later processing into other types of data. The most common example of this is technical indicators that are frequently used as a source of data in systems for algorithmic trading.We also find a number of machine learning techniques that have been demonstrated by previous publications to be applicable for algorithmic trading. Publications show that a machine learning technique called SVM (support vector machine) can be applied on technical indicators as well as for analysis of news headlines. We also find publications that demonstrate the application of two types of neural networks, classification and regression network. These are used in order to generate trade signals in an algorithmic trading system. In our study we also find an application of evolutionary machine learning which is used to approximate an optimal solution to the order execution problem. Moreover, we also discuss a financial incentive that disadvantage academic openness and the publications of new discoveries in the relevant area of research. This financial incentive exists because advantageous results may be financially beneficial to withhold.
Hjalmarsson, Martin, and Mikael Björkman. "Bedömning av fakturor med hjälp av maskininlärning." Thesis, Örebro universitet, Institutionen för naturvetenskap och teknik, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:oru:diva-64835.
Full textToday, companies can sell their invoices to a third party in order to to quickly capitalize them. This is called factoring. For the financial institute which serve as the third party, the purchase of an invoice infers a certain risk in case the invoice is not paid, a risk the financial institute would like to minimize. Aros Kapital is a financial institute that offers factoring as one of their services. This project at Aros Kapital evaluated the possibility of using machine learning to determine whether or not an invoice will be good investment for the financial institute. If the machine learning algorithm performs better than manual handling and by minimizing credit losses and buying more invoices this could lead to an increase in profit for Aros. Four machine learning algorithms have been compared: decision trees, random forest, Adaboost and deep neural network. Beyond the comparison between the four algorithms, the algorithms were also compared with Aros actual decision and Aros current rule engine solution. The results show that random forest is the best performing algorithm and it also shows a slight improvement on performance compared to Aros actual decision, random forest got an F1- core of 0.35 and Aros 0.22.
Lerdell, André, and Simon Shadman. "Maskininlärning inom kundanalys : Prediktion av kundbeteende inom energibranchen." Thesis, Uppsala universitet, Avdelningen för systemteknik, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-376295.
Full textPayerl, Anders. "Tolka musiktecken från bilder : Optisk musikigenkänning med maskininlärning." Thesis, Mittuniversitetet, Avdelningen för informationssystem och -teknologi, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:miun:diva-34063.
Full textMålet med projektet har varit att undersöka om det går att applicera maskininlärning utan att ha kunskaper av maskininlärning och matematik sedan tidigare samt om detta är möjligt också undersöka om det går att använda maskininlärning för att tolka en bild av nottecken för musik. Detta användes sedan för att jämföra dess förmåga att upptäcka noter med ett redan existerande program kallat Audiveris. Metoden som användes var att bilder märktes med information om dess innehåll och sedan användes dessa bilder för att träna en maskininlärningsmodell att tolka bilder av nottecken. Sedan gjordes en manuell jämförelse av Audiveris resultat samt resultatet från maskininlärningsmodellen efter att den tränats på de nya notbilderna. Resultatet blev en modell som sedan användes vid jämförelsen med Audiveris. Den jämförelsen resulterade i att Audiveris visade sig bättre än den nytränade modellen då Audiveris hittade nästan 100% av noterna på bilden medan den nya modellen bara hittade cirka 33.3%. Orsaken till den stora skillnaden på upptäckande av nottecken berodde antagligen till stor del på två saker: den första att Audiveris utvecklats under många år och den andra att träningsdatat som användes till den nya modellen inte var tillräckligt varierat och komplicerat. För att vidareutveckla den nya modellens färdighet skulle framför allt mer träningsdata behöva användas vid träningsmomentet. Även då behövde man se till att variationen av material och dess svårighetsgrad blev mer varierat än vid grundmaterialet.
Clase, Christian. "Maskininlärning och bildtolkning för ökad tillförlitlighet i strömavtagarlarm." Thesis, Högskolan i Gävle, Avdelningen för elektronik, matematik och naturvetenskap, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:hig:diva-27766.
Full textAlemayehu, Brook, and Fredrik Johnsons. "Maskininlärning inom kommersiella fastigheter : Prediktion av framtida hyresvakanser." Thesis, Uppsala universitet, Avdelningen för systemteknik, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-357245.
Full textLiu, Julia, and Linnéa Lindahl. "Prediktion av efterfrågan i filmbranschen baserat på maskininlärning." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-235719.
Full textMaskininlärning är en central teknik i datadrivet beslutsfattande. I den här rapporten utreds maskininlärning isammanhanget av efterfrågeprediktion i filmbranschen från biografers perspektiv. Närmare bestämt undersöks det i vilken utsträckningtekniken kan bistå uppskattning av publikintresse i termer av intäkter vad gäller osläppta filmer hos biografer. Tremaskininlärningsmodeller implementeras i syfte att göra en prognos på kumulativa intäktsnivåer under premiärhelgen för filmer vilkahade premiär 2010-2017 i Sverige. Prognostiseringen baseras på varierande attribut som sträcker sig från publik användargenererad data på nätet till filmspecifika variabler så som produktionsbudget och uppsättning av skådespelare. De erhållna resultaten visar att valen av attribut och modeller inte var optimala på den svenska marknaden då erhållna precisionsmått från modellerna antog låga värden, med relevanta underliggande skäl.
Petersson, Linda, and Magdalena Äng. "Etisk problematik vid implementering av maskininlärning i sjukvården." Thesis, Uppsala universitet, Institutionen för informatik och media, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-413702.
Full textNilsson, Gustav, and William Takolander. "Smarta rekommendationer : Rekommendationer på webbsidor framtagna av maskininlärning." Thesis, Mittuniversitetet, Institutionen för informationssystem och –teknologi, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:miun:diva-39345.
Full textIn today's society machine learning is a growing method to solve certain problems faced by companies worldwide. Many companies have mountains of stored data that are not being utilised. This data can be used in numerous ways to make improvements within these companies. One of the ways is machine learning, it is used more and more these days to generate recommendations. This project's purpose is to make a proof of concept of a machine learning model capable of giving recommendations based on historical data. This proof of concept will serve as guidelines to Centrala Studiestödsnämnden (CSN) in how they should approach machine learning as an alternative to manual recommendations. This is achieved by determining what data is to be used, understanding the data selected and then picking an algorithm suitable for that data. Then the algorithms will be used to create machine learned models which will be tested in various ways to see which works best for the task at hand. Two models are created with different algorithms that both fit the purpose. The models are tested through practical and theoretical tests. The results show that the algorithms are similar in which predicted recommendations they give but have slight variation.
Assarsson, Fabian, and Gustav Kjellin. "Maskininlärning som möjligt planeringsverktyg för mindre serviceföretag : En studie i maskininlärning för verksamhetsplanering inom hissbranschen hos fallstudieföretaget S:t Eriks Hiss AB." Thesis, KTH, Skolan för datavetenskap och kommunikation (CSC), 2016. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-189468.
Full textIn this report we show that Machine Learning algorithms from readily available libraries can be useful tools when analyzing company data. We suggest a new metric for analyzing the frequency of unplanned maintenance calls with an elevator maintenance case study, and also provide the company in question, S:t Eriks Hiss, with a K-Means clustered suggestion of geographical customer partitioning. We provide a brief discussion on whether the results from Anderson et al (2009) and Blakeley et al. (2003) could provide an analytical framework for companies in the urban maintenance industry. A weak negative correlation between profitability and our suggested metric indicate that so is the case and makes for interesting further enquiry.
Fredén, Gustaf, and Erik Persson. "Maskininlärning som verktyg för att förbättra planering i produktion." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-239630.
Full textProduction planning for products with high seasonality can be difficult, as demand can increase rapidly, and it can be difficult to predict exactly when this increase will happen. Propane gas is such a product, where demand is significantly higher during summer than during winter. This study is conducted for AGA, the market leader for propane in the Nordic region, with the purpose to explore if machine learning can be used to improve their production planning. Different regression models where used to execute the study, with the result that none of them could be implemented with high enough proficiency to be used to predict future demand of propane and aid the production planning. The reason that the implementations failed was that too few input parameters where used for the machine learning models, and that those parameters that were used had too low correlation.
Normark, Tim, and Isabella Andrén. "Användande av maskininlärning vid skapande och utvärdering av kursplaner." Thesis, Malmö universitet, Fakulteten för teknik och samhälle (TS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:mau:diva-20291.
Full textThe study investigates whether machine learning can be used to classify syllabuses well enough to possibly be used as an aid in the process of assessing new syllabuses. The study also examines whether with unsupervised learning one can see clear differences regarding the semantics of the different parts of the learning objectives in syllabuses.An experimental research approach has been used and the approach is inspired by the various steps in the CRISP data mining process. In practice, this meant that a number of applications were built to retrieve and process data in the form of syllabuses and then machine learning models were created that were used experimentally to answer the research questions formulated.The results of the study have shown that the classification models are not sufficient enough to be used as an aid in the process of assessing new syllabuses. The model’s assessment of whether a syllabus is approved or not approved had a high accuracy on the collected data that was used to train and test the models. To test how well a model could be used in a real context, the classification of another set of syllabuses was compared to the evaluation of the same syllabuses by a teacher of Malmö University. In this context the result of the model’s and the teacher’s evaluation matched very poorly. The results regarding the use of unsupervised learning in relation to the learning objectives in syllabuses showed that this machine learning strategy appears to be effective for this purpose. The results here showed that there are no clear differences between the classification of the learning objectives in the syllabuses used as data for the models
Bojs, Robert, and Benny Feng. "Övervakad maskininlärning för att identifiera nya kunder på energimarknaden." Thesis, KTH, Skolan för datavetenskap och kommunikation (CSC), 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-210782.
Full textDet här arbetet utforskar alternativa tillvägagångssätt för för mindre aktörer på energimarknaden att identifiera nya potentiella kunder, baserat på publikt tillgänglig data som analyseras med hjälp av maskininlärningsalgoritmer. På senare år har pris ansetts vara den faktor som påverkar val av leverantör mest. Eftersom större aktörer på marknaden kan utnyttja skalfördelar kan de pressa priserna hårt, medans mindre aktörer måste finna andra vägar att vinna nya kunder. Maskininlärningsalgoritmerna i den här uppsatsen kommer att använda försäljningsdata från ett litet energibolag, som bedriver verksamhet i Sverige, med målet att hitta ett mönster mellan existerande kunder och deras demografiska data. Genom att förskaffa sig djupare kunskap om vad som differentierar kunder kan energibolaget förbättra sin försäljning. På grund av en förhållandevis liten mängd kunddata och brist på publik data gick det inte att hitta ett betydande samband mellan kunderna och deras demografiska data. Resultaten utgör dock en bra grund för fortsatt forskning då resultaten blir mer pålitliga då mer kunddata införskaffas, vilket blir en naturlig följd av att energibolagets försäljning fortsätter utvecklas.
Saari, Lukas, and Emil Mårtensson. "Oövervakad maskininlärning för att upptäcka bottar i online-tävlingar." Thesis, KTH, Skolan för datavetenskap och kommunikation (CSC), 2016. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-209821.
Full textDigital marketing is a fast-growing market and its actors are constantlylooking for innovative and new ways of marketing. In this paper, an actoron this market called Adoveo will be studied. Their specialization and valueproposition is to include a competition part in their advertisement campaigns,giving its participators the possibility to win a prize. What could turn out to beproblematic is that the prizes are not rewarded to human contestants, insteadgoing to a bot that can participate in the competition with unreasonably goodresults. The purpose of this paper is to try to separate bots from human contestantswith the data provided from Adoveo. To that end, two unsupervised machinelearning algorithms were implemented to cluster the data points, GaussianMixture Model and K-Means. The result was an uninterpretable cluster structurefrom which there was no reliable identification of bot-like and human-likebehaviour to be made. The reason behind this was twofold, the design of thecompetition and a lack of decisive attributes in the data. Recommendationswere provided to how both of these issues could be rectified.Finally, an analysis was provided on the business value of bot-securingcompetitions and the value it gives to the company. The analysis showed thatthe business value of bot-securing competitions would be beneficial, becauseit would give a competitive advantage against competitors and also improvebusiness with advertisers and consumers.
Öbom, Anton, and Adrian Bratteby. "Maskininlärning för att förutspå churn baserat på diskontinuerlig beteendedata." Thesis, KTH, Skolan för datavetenskap och kommunikation (CSC), 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-210546.
Full textDenna rapport handlar om att utforska fälten maskininlärning och digital marknadsföring, genom att använda maskininlärning som ett redskap för att förutspå churn i en typ av företag med diskontinuerlig beteendedata. För att förutspå churn finns relativt simpla "out of the box"-modeller, som support vector machines och random forests, som används för att nå acceptabla resultat. För att nå liknande resultat som i arbeten där churn utförs på kontinuerlig beteendedata konstaterar denna rapport att framtida arbeten forska på vilka utvärderingsmetriker som är mest lämpade. I rapporten presenteras också hur dessa upptäckter kan kommersialiseras och hur företaget Sellpy kan tjäna på att förutspå churn.
Bile, Excell Linus. "Fundamentala utmaningar med maskininlärning : Identifikation av ansiktsmask på bild." Thesis, Mittuniversitetet, Institutionen för informationssystem och –teknologi, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:miun:diva-42962.
Full textMachine learning is a widely used technique, which can be used for image recognition. The aim of this project is to get a basic understanding for how machine learning operates, including the user’s required technical prior knowledge as well as the challenges that exist within a self-learning system. This was examined by creating and optimizing a system that identifies whether a person in a picture is wearing a face mask or not. The main focus of the project has been on collecting and managing data, but most importantly on optimizing hyper parameters. This was executed by collecting information to achieve basic understanding of the topic. Then the system was trained, validated and tested. The system was optimized by application of various hyper parameters to show the user how they affect the result. This was executed in Keras and visualized in MatPlotlib. The result showed that one challenge in a self-learning system is to reduce the risk of overfitting, which is why application of the hyper parameter dropout was important. The challenge in using machine learning seemed to be that many hyper parameters can affect the result, and understanding what, how and why a result is the way it is can be difficult for the user. Despite this, a system that could interpret whether a person in a picture was wearing a face mask or not was created and optimized in a sufficient way regarding the amount of data, time and previous knowledge available. This emphasizes the utility of machine learning both in this and other areas.
Engblom, Emil. "Nyttiggörande avmaskininlärningsmodeller i verksamheten : Ökad metadatakvalitet med stöd från maskininlärning." Thesis, Uppsala universitet, Data- och systemvetenskap, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-437156.
Full textLjung, Mikael, and Linnea Månsson. "Klassificering av engagemangsnivå hos en samtalsdeltagare med hjälp av maskininlärning." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-262046.
Full textArbetet som presenteras i den här studien grundar sig i det långsiktiga målet att utveckla en social robot som kan vara med och leda samtalssessioner på ett språkcafé. I detalj har studien undersökt om det går att klassificera engagemang hos en samtalsdeltagare utifrån dess ansiktsuttryck och blickriktning – två faktorer som tidigare studier visat sig vara centrala för människans engagemang. För att utföra bedömningen har mjukvaran Openface extraherat nämnda parametrar från en tidigare fältstudie vilka sedan har processats med maskininlärningsmodellen Support Vector Machine. Efter gedigna försök att finna optimala värden på hyperparametrar till modellen lyckades den slutligen predicera engagemang på en tregradig skala med 54,5% accuracy. Vidare har studien också undersökt potentialen för det nya teknologiska paradigmet som den sociala roboten utgör. Potentialen har analyserats med utgångspunkt i Dosis fyra dimensioner: teknologiska möjligheter, möjliga vinster från innovation, kumulativ höjd på teknologiska framsteg och egenskaper i kunskapsbasen. Analysen klargör att paradigmet har förutsättningar att revolutionera ett flertal industrier till följd av dess teknologiska möjligheter och världsomfattande intressenter, men står också inför utmaningar i form av tekniska och etiska svårigheter.
Berg, Tom, and Sergej Engström. "Maskininlärning som medel för att förutsäga NBA:s mest värdefulla spelare." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-272104.
Full textDenna studie har utförts inom områdena industriell ekonomi och datateknik. Fokus har legat på maskininlärning och med vilken precision det kan användas för att förutspå mottagare av idrottsutmärkelser, baserat på deras individuella prestationer. I arbetet har olika maskininlärningsalgoritmer använts och utvärderats för att förutsäga vem som utses till den mest värdefulla spelaren i National Basketball Association. Kvantitativa data har samlats in från nästan 20 säsonger för att lägga grunden för träningen av algoritmerna. Features har valts ut och hyperparametrar har ställts in för att optimera förutsägelserna. Algoritmerna har sedan utvärderats och jämförts för att kunna maximera precisionen. Arbetet resulterade i förutsägelser med R2-scores på över 80% samt att varje metod förutsåg fler än hälften av alla MVPs.
Schnackenburg, Ellen Cecilia, and Karl Leife. "Maskininlärning i fastighetsbranschen : Prediktion av felanmälningar gällande inomhusklimat baserat på sensordata." Thesis, Uppsala universitet, Avdelningen för systemteknik, 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-338579.
Full textKirik, Engin. "Tolkning av handskrivna siffror i formulär : Betydelsen av datauppsättningens storlek vid maskininlärning." Thesis, Mittuniversitetet, Institutionen för informationssystem och –teknologi, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:miun:diva-41291.
Full textThe research in this study has been to extract how important the size of the dataset is for the impact on results within object recognition. The research was implemented in training a model in computer vision that should be able to identify and convert handwritten numbers from physical forms to digitized format. Two different frameworks called TensorFlow and PyTorch were used for this process. The process was trained in two different environments, one model was trained in the CPU environment and the other in the Google Cloud GPU environment. The idea of the study is to improve results from previous degree projects and further research to expand the development extra by creating a model that identifies and digitizes several handwritten numbers simultaneously on a complete form, which will continue to be able to help and be used in the future for applications that sums up points on a form using a mobile camera for recognition. The project showed a result of an error-free recognition of several numbers at the same time, when the data set was constantly expanded. Results around individual numbers managed to identify all numbers from 0 to 9 with both the TensorFlow and PyTorch frameworks.
Jacobsson, Marcus, and Viktor Inkapööl. "Prediktion av optimal tidpunkt för köp av flygbiljetter med hjälp av maskininlärning." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-281767.
Full textArbetet presenterat i studien är baserat på målet att sänka konsumentkostnader relaterat till köp av flygresor. Mer specifikt har studien undersökt huruvida det är möjligt att predicera optimala köpbeslut för specifika flygrutter med hjälp av maskininlärningsmodeller tränade på grundläggande data innehållande endast information om pris och sökdatum för varje givet avresedatum. Modellerna baserades på Random Forest Classifier och tränades på sökdata upp till 90 dagar före avresa för varje avresedag i juli 2016–2018, och testades på likadan data för 2019. Efter förberedelse av data och tuning av hyperparametrar lyckades modellerna med en träffsäkerhet på 88% respektive 84% predicera optimalt köp för rutterna Stockholm-Mallorca respektive Stockholm-Bangkok. Baserat på antagande om att antalet sökningar korrelerar med efterfrågan och vidare faktiska köp, beräknade studien att den genomsnittliga förväntade besparingen per biljett vid användning av modeller på de undersökta rutterna till 21% respektive 17%. Vidare undersökte studien hur en affärsmodell för prisjämförelse kan omformas för att inkorporera resultaten. Ramverkat som användes för detta var Business Model Canvas och mynnade ut i en rekommendation av implementering av en premiumtjänst genom vilken användare ges information biljett ska köpas eller ej vid en given sökning.
Marchal, Jakob, and Mathias Andreasen Andreasen. "Autopositionering för röntgensystem." Thesis, Linnéuniversitetet, Institutionen för datavetenskap (DV), 2014. http://urn.kb.se/resolve?urn=urn:nbn:se:lnu:diva-42742.
Full textPersson, Masud Alexander, and Viktor Olsson. "Cyklisters upplevda otrygghet i urban miljö – En studie med klusteranalys." Thesis, Malmö universitet, Fakulteten för teknik och samhälle (TS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:mau:diva-20207.
Full textThe municipality of Lund strives to be a leading municipality in Sweden regarding bicycle usage as a means of transportation. With the machine learning type cluster analysis, we want to examine the possibility of analysing data generated by cyclist in Lund in order to understand cyclists perceived insecurity in traffic. The purpose of performing cluster analysis was to more cost efficiently further enhance a safer environment for cyclists. We perform our analysis based on the K-means algorithm and further develop two different methods. The first method is based on Euclidian distance and the second method is based on the amount of datapoints in a given cluster. These methods control for the size of a cluster in order to match the geographical space we are working with such as roads and crossings. The Euclidian method generates clusters based on size in meters and the other method generates clusters based on amount of datapoints. In our experiment the result shows that the Euclidian method is more suited for cluster analysis. We also believe that the usability of cluster analysis with our methods isn't sufficient in order for us to believe that cluster analysis is more usable than none machine learning analysis. By adding additional variables to the data collection and comparing cluster analyses over time we believe that cluster analysis could be of more value.
Melander, Gustav, and Robin Wänlund. "Maskininlärning som medel för att betygsätta samtal med språklärande syfte mellan robot och människa." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-259552.
Full textThe Swedish company Furhat Robotic have created a robot called Furhat, which is able to interact with humans in a language café setting. The purpose of the robot led conversation is for the participants to develop their language skills. After the conversation the humans will answer a survey about what they thought about the conversation with Furhat. A question that has arisen from this is if it is possible to predict the survey answers based on just the conversation. The purpose of this paper is to analyze if it is possible to quantify the conversations linked to the survey answers, and by doing so be able to predict the answers in new conversations with a machine learning approach. The data set being used was obtained from an earlier study in Collaborative Robot Assisted Language Learning. The result returned a RMSE that was greater than the variance of the average conversation score which indicates that the model is not very effective. However, it excelled in some predictions trying to give scores to each separate survey answer, indicating that the model could be used for certain question formulations.
Appelquist, Niklas, and Emelia Karlsson. "Kan en bättre prediktion uppnås genom en kategorispecifik modell? : Teknologiprojekt på Kickstarter och maskininlärning." Thesis, Uppsala universitet, Institutionen för informatik och media, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-413736.
Full textCrowdfunding is used to collect money via internet for potential projects through a large number of backers which contribute with small pledges. Kickstarter is one of the largest crowdfunding platforms today. Despite the big interest in crowdfunding a lot of launched campaigns fail to reach their goal and projects of the category technology shows the largest rate of failure on Kickstarter. Therefore, it is important to be able to predict which campaigns are likely to succeed or fail. This thesis aims to explore the possibility of reaching a higher accuracy when predicting the success of launched projects with machine learning with a smaller amount of category-specific data. The data consists om 192 548 launched projects on Kickstarter and has been collected through Kaggle.com. Two models of the type Random Forest has been developed where one model has been trained with general data over all projects and one model has been trained with category specific data over technology projects. The results show that the technology model show a higher accuracy rate with 68,37 % compared to the reference model with 68,00 %.
Theo, Sobczak. "Maskininlärning och fallklassificering med MEMS-accelerometer : En studie i fallklassificering med artificiella neurala nätverk." Thesis, Högskolan i Gävle, Elektronik, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:hig:diva-34013.
Full textBerglund, Frida, and Vendela Talenti. "Maskin eller läkare? En studie om individens attityd till användning av vårdapplikationer med maskininlärning." Thesis, Malmö universitet, Fakulteten för teknik och samhälle (TS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:mau:diva-20881.
Full textThis study aims to research on individuals’ general attitudes towards healthcare applicationsthat use machine learning. The data collection has taken place through both qualitative andquantitative methods as a complement to each other. The methods include a questionnairesurvey, two focus groups based on scenario-based design. The theory is based on research in thedigitalisation of healthcare, including machine learning and mHealth, which is based andsupports the investigation. The theory of attitudes and confidence in the digitalisation of carealso forms the basis for the study.The conclusion shows that there is a correlation between high awareness and positive attitudewhen it comes to the use of healthcare applications with machine learning. The general attitudetowards a diagnosis from machine learning is negative since most people prefer to get adiagnosis mediated by a doctor. The study indicates that this may be because the patients seekempathy from the healthcare system, which artificial intelligence lacks of. Trust towards ahealthcare application is based primarily on the reputation of it, but also in which company ororganization that is behind it. The respondents in the survey are positive about contributing withtheir personal data to a healthcare application if it leads to a prevention of a disease. The studyalso gives an indication that there is a fear of what happens with private health data.
Ekeberg, Lukas, and Alexander Fahnehjelm. "Maskininlärning som verktyg för att extrahera information om attribut kring bostadsannonser i syfte att maximera försäljningspris." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-240401.
Full textDen svenska bostadsmarknaden har blivit alltmer digitaliserad under det senaste årtiondet med nuvarande praxis att säljaren publicerar sin bostadsannons online. En fråga som uppstår är hur en säljare kan optimera sin annons för att maximera budpremie. Denna studie analyserar tre maskininlärningsmetoder för att lösa detta problem: Linear Regression, Decision Tree Regressor och Random Forest Regressor. Syftet är att utvinna information om de signifikanta attribut som påverkar budpremien. Det dataset som använts innehåller lägenheter som såldes under åren 2014-2018 i Stockholmsområdet Östermalm / Djurgården. Modellerna som togs fram uppnådde ett R²-värde på approximativt 0.26 och Mean Absolute Error på approximativt 0.06. Signifikant information kunde extraheras from modellerna trots att de inte var exakta i att förutspå budpremien. Sammanfattningsvis skapar ett stort antal visningar och en publicering i april de bästa förutsättningarna för att uppnå en hög budpremie. Säljaren ska försöka hålla antal dagar sedan publicering under 15.5 dagar och undvika att publicera på tisdagar.
Larsson, Olsson Simon. "Användandet av algoritmer inom investeringar kopplat till OMX30 : Tillämpning av maskininlärning inom portföljhantering: En K-Betydelsemetod." Thesis, Södertörns högskola, Institutionen för samhällsvetenskaper, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:sh:diva-43906.
Full textJohan, Pedersen, and Alfredsson Rebecka. "Prognostisering med hjälp av maskininlärning." Thesis, Karlstads universitet, Institutionen för matematik och datavetenskap (from 2013), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kau:diva-78151.
Full textHällgren, Clara, and Alexander Kristiansson. "Maskininlärning för automatisk extrahering av citat från recensioner : Med användning av BERT, Inter-Sentence Transformer och artificiella neuronnätverk." Thesis, KTH, Hälsoinformatik och logistik, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-296565.
Full textTo choose a number of sentences from a movie review to use as a quote can be time consuming if done manually. This thesis evaluates supervised machine learning models to create a prototype that automatically can choose such quotes. The thesis chose, based on a literature study, two models to implement and evaluate on data consisting of movie reviews and their respective corresponding manually chosen quotes. Out of the thesis two implemented models, BERT with Inter-Sentence Transformer and BERT with an artificial neural network, the latter showed marginally better results. The models were evaluated with ROUGE and was compared with state-of-the-art models regarding automatic text summarization. The conclusion is that the models that were evaluated do not perform well enough for the problem to motivate full deployment without further development efforts. However, the results show that there is potential that the evaluated methods can partially replace manual labour when choosing quotes.