Recent neural network advances have lead to models that can be used for a variety of image classification tasks, useful for many of today’s media technology applications. In this paper, I train hallmark neural network architectures on a newly collected vehicle image dataset to do both coarse- and fine-grained classification of vehicle type. The results show that the neural networks can learn to distinguish both between many very different and between a few very similar classes, reaching accuracies of 50.8% accuracy on 28 classes and 61.5% in the most challenging 5, despite noisy images and labeling of the dataset. / Nya neurala nätverksframsteg har lett till modeller som kan användas för en mängd olika bildklasseringsuppgifter, och är därför användbara många av dagens medietekniska applikationer. I detta projektet tränar jag moderna neurala nätverksarkitekturer på en nyuppsamlad bilbild-datasats för att göra både grov- och finkornad klassificering av fordonstyp. Resultaten visar att neurala nätverk kan lära sig att skilja mellan många mycket olika bilklasser,  och även mellan några mycket liknande klasser. Mina bästa modeller nådde 50,8% träffsäkerhet vid 28 klasser och 61,5% på de mest utmanande 5, trots brusiga bilder och manuell klassificering av datasetet.

As imaging systems become ubiquitous, the ability to recognize human actions is becoming increasingly important. Just as in the object detection and recognition literature, action recognition can be roughly divided into classification tasks, where the goal is to classify a video according to the action depicted in the video, and detection tasks, where the goal is to detect and localize a human performing a particular action. A growing literature is demonstrating the benefits of localizing discriminative sub-regions of images and videos when performing recognition tasks. In this thesis, we address the action detection and recognition problems. Action detection in video is a particularly difficult problem because actions must not only be recognized correctly, but must also be localized in the 3D spatio-temporal volume. We introduce a technique that transforms the 3D localization problem into a series of 2D detection tasks. This is accomplished by dividing the video into overlapping segments, then representing each segment with a 2D video projection. The advantage of the 2D projection is that it makes it convenient to apply the best techniques from object detection to the action detection problem. We also introduce a novel, straightforward method for searching the 2D projections to localize actions, termed TwoPoint Subwindow Search (TPSS). Finally, we show how to connect the local detections in time using a chaining algorithm to identify the entire extent of the action. Our experiments show that video projection outperforms the latest results on action detection in a direct comparison. Second, we present a probabilistic model learning to identify discriminative regions in videos from weakly-supervised data where each video clip is only assigned a label describing what action is present in the frame or clip. While our first system requires every action to be manually outlined in every frame of the video, this second system only requires that the video be given a single highlevel tag. From this data, the system is able to identify discriminative regions that correspond well iii to the regions containing the actual actions. Our experiments on both the MSR Action Dataset II and UCF Sports Dataset show that the localizations produced by this weakly supervised system are comparable in quality to localizations produced by systems that require each frame to be manually annotated. This system is able to detect actions in both 1) non-temporally segmented action videos and 2) recognition tasks where a single label is assigned to the clip. We also demonstrate the action recognition performance of our method on two complex datasets, i.e. HMDB and UCF101. Third, we extend our weakly-supervised framework by replacing the recognition stage with a twostage neural network and apply dropout for preventing overfitting of the parameters on the training data. Dropout technique has been recently introduced to prevent overfitting of the parameters in deep neural networks and it has been applied successfully to object recognition problem. To our knowledge, this is the first system using dropout for action recognition problem. We demonstrate that using dropout improves the action recognition accuracies on HMDB and UCF101 datasets.

The Greek Parliament is a critical institution for the Greek Democracy, where important decisions are made that affect the lives of millions of people. It consists of representatives from different political parties, and each party has a unique political ideology, stance, and agenda. The proposed research aims to automatically classify parliamentary proceedings to their respective political parties based on the content of their speeches, debates, and discussions. The goal of this research is to assess the feasibility of classifying Greek parliament proceedings to their respective political party via machine learning and neural network algorithms. By using machine learning algorithms and neural networks, the system can learn from large amounts of data and make accurate predictions about the category of a given proceeding. One possible approach is to use supervised learning algorithms, where the system is trained on a dataset of parliamentary proceedings labeled with the respective political parties. The machine learning algorithms can then learn the underlying patterns and features in the text data and accurately classify new proceedings to their respective parties. Another potential approach is to use deep learning neural networks, such as recurrent neural networks (RNNs), to classify the proceedings. These networks can be trained on large amounts of labeled data and can learn the complex relationships between the text features and political parties. The results of this research can be used to gain insights into the political landscape and the positions of different parties on various issues. The ability to automatically classify parliamentary proceedings to their political parties can also aid in political analysis, including tracking the voting patterns of different parties and their representatives and generally the potential revolutionization of social and human sciences is existent. Moreover, the proposed research can have implications for policy-making and governance. By analyzing the proceedings and identifying the political parties' positions and priorities, policymakers can better understand the political landscape and craft policies that align with the values and priorities of different parties. In conclusion, the classification of parliament proceedings, in our case Greek, to their political parties via NLP with machine learning algorithms is a promising research topic that has potential applications in political analysis and decision-making. The ability to automatically classify parliamentary proceedings to their respective parties can enhance transparency and accountability in the democratic system and aid in policy-making and governance.

Identifying Low-head dams (LHD) and creating an inventory become a priority as fatalities continue to occur at these structures. Because obstruction inventories do not specifically identify LHDs, and they are not assigned a hazard classification, there is not an official inventory of LHD. However, there is a multi-agency taskforce that is creating an inventory of LHD. All efforts have been performed by manually identifying LHD on Google Earth Pro (GE Pro). The purpose of this paper is to assess whether a machine learning approach can accelerate the national inventory. We used a machine learning approach to implement a high-resolution remote sensing data and a Convolutional Neural Network (CNN) architecture. The model achieved 76% accuracy on identifying LHD (true positive) and 95% accuracy identifying NLHD (true negative) on the validation set. We deployed the trained model into the National Hydrologic Geospatial Fabric (Hydrofabric) flowlines on the Provo River watershed. The results showed a high number of false positives and low accuracy in identifying LHD due to the mismatch between Hydrofabric flowlines and actual waterways. We recommend improving the accuracy of the Hydrofabric waterway tracing algorithms to increase the percentage of correctly classified LHD.

Careful feature engineering is an important factor of artificial intelligence for games. In this thesis I investigate the benefit of delegating the engineering efforts to the model rather than the features, using the board game Othello as a case study. Convolutional neural networks of varying depths are trained to play in a human-like manner by learning to predict actions from tournaments. My main result is that using a raw board state representation, a network can be trained to achieve 57.4% prediction accuracy on a test set, surpassing previous state-of-the-art in this task.  The accuracy is increased to 58.3% by adding several common handcrafted features as input to the network but at the cost of more than half again as much the computation time. / Noggrann funktionsteknik är en viktig faktor för artificiell intelligens för spel. I dennaavhandling undersöker jag fördelarna med att delegera teknikarbetet till modellen i ställetför de funktioner, som använder brädspelet Othello som en fallstudie. Konvolutionellaneurala nätverk av varierande djup är utbildade att spela på ett mänskligt sätt genom attlära sig att förutsäga handlingar från turneringar. Mitt främsta resultat är att ett nätverkkan utbildas för att uppnå 57,4% prediktionsnoggrannhet på en testuppsättning, vilketöverträffar tidigare toppmoderna i den här uppgiften. Noggrannheten ökar till 58.3% genomatt lägga till flera vanliga handgjorda funktioner som inmatning till nätverket, tillkostnaden för mer än hälften så mycket beräknatid.

Recent years have seen multiple machine-learning research projects concerning agents in video games. Yet, there is a disjoint between this academic research and the video game industry, evidenced by the fact that game developers still hesitate to use neural networks (NN) due to lack of clarity and control. Particularly for denizens, which are agents that take specific roles and have highly specialized purposes. Many denizens share features that could be exploited to reduce the hardship of training different types of denizens. A Cooperative Modular Neural Network (CMNN) seeks to provide more clarity and control than a monolithic neural network (Mono-NN) by breaking down the problem into specialist modules that exploit common denizen features and fuse them via a main network. The objective is to compare the CMNN and the Mono-NN in technical performance, and to compare the player satisfaction of playing against the two approaches in the same video game, Star Fetchers. The game was chosen because it belongs to the established genre of two-dimensional platforming games, providing a simple context. All NNs were implemented using the library TorchSharp. The approaches were compared on frame time, memory usage, and training time. A User Study of 58 participants' opinions regarding engagement and denizen movement was conducted and the results were analyzed for any statistical significance. The CMNN approach was shown to perform worse in frame time and memory usage. However, through parallelization of the modules, and by sharing modules between CMNNs, the gap can be bridged slightly. The training time was shown to be worse for the CMNN compared to the Mono-NN. Backward propagation, however, was faster for the CMNN, counterbalancing the time lost during forward propagation at shorter episode lengths. The CMNN also produces a minimum viable denizen in fewer epochs, significantly reducing the real-time spent training the denizen. The results of the User Study was inconclusive due to statistical insignificance. The CMNN is a viable competitor to Mono-NNs, at least in some aspects. Training is still costly in terms of time and effort and the complexity concerning hyperparameters and intelligent choice of reward function remains. However, the modules provide out-of-the-box networks that can be reused. More work within the area of cooperative modular methods is needed before the video game industry has any reason to make the change over from other time-proven methods. / De senaste åren har flera maskininlärningsforskningsprojekt om agenter i datorspel genomförts. Trots detta finns en klyfta mellan denna akademiska forskning och datorspelsindustrin. Detta tydliggörs av det faktum att spelutvecklare fortfarande tvekar att använda neurala nätverk på grund av bristande klarhet och kontroll. Detta gäller särskilt "invånare", agenter som har specifika roller och specialiserade syften. Många invånare delar egenskaper som skulle kunna utnyttjas för att minska svårigheten med att träna olika typer av invånare. Ett Kooperativt Modulärt Neuralt Nätverk (CMNN) strävar efter att ge mer klarhet och kontroll än ett monolitiskt neuralt nätverk (Mono-NN) genom att bryta ned problemet i specialiserade moduler som utnyttjar gemensamma egenskaper hos invånare och förenar dem via ett huvudnätverk. Syftet är att jämföra ett CMNN och ett Mono-NN i teknisk prestanda, och att jämföra användarupplevelsen då användaren spelar mot de två metoderna i samma datorspel, Star Fetchers. Spelet valdes då det tillhör den väletablerade genren av två-dimensionella plattformsspel, vilket ger en simpel kontext för arbetet. Båda neurala nätverken implementerades med biblioteket TorchSharp. Nätverken jämfördes med avseende på tid per bild, minnesanvändning och träningstid. En användarstudie samlade åsikter från 58 deltagare angående spelarens engagemang och invånarnas rörelse, vilket analyserades för eventuella statistiska signifikanser. CMNN presterade sämre med tanke på tid per bild och minnesanvändning. Dock, genom parallellisering och delning av moduler mellan flera CMNN, kan klyftan mellan dem minskas. Träningstiden visade sig vara sämre för CMNN jämfört med Mono-NN. Bakåtpropagering var dock snabbare med CMNN, vilket kompenserar för den tid som förloras under framåtpropagering vid kortare episodlängder. CMNN producerar också en acceptabel invånare på färre epoker, vilket markant minskar den verkliga tiden som spenderas på att träna invånare. Resultaten från användarstudien var inte övertygande på grund av brist på statistisk signifikans. CMNN är ett bra alternativ till Mono-NN, åtminstone med tanke på vissa aspekter. Träningen är fortfarande resurskrävande i form av tid och ansträngning och komplexiteten kring hyperparametrar och intelligent val av belöningsfunktion består. Modulerna tillhandahåller dock färdiga nätverk som kan återanvändas. Det krävs i framtiden mer arbete inom kooperativa och modulära metoder innan datorspelsindustrin har någon anledning att byta över från andra, beprövade metoder.

Background: Cyberbullying involves the use of digital technologies to harass, humiliate, or threaten individuals or groups. This form of bullying can occur on various platforms such as social media, messaging apps, gaming platforms, and mobile phones. With the outbreak of covid-19, there was a drastic increase in utilization of social media. And this upsurge was coupled with cyberbullying, making it a pressing issue that needs to be addressed. Sentiment analysis involves identifying and categorizing emotions and opinions expressed in text data using natural language processing and machine learning techniques. SVM is a machine learning algorithm that has been widely used for sentiment analysis due to its accuracy and efficiency. Objectives: The main objective of this study is to use SVM for sentiment analysis of cyberbullying tweets and evaluate its performance. The study aimed to determine the feasibility of using SVM for sentiment analysis and to assess its accuracy in detecting cyberbullying. Methods: The quantitative research method is used in this thesis, and data is analyzed using statistical analysis. The data set is from Kaggle and includes data about cyberbullying tweets. The collected data is preprocessed and used to train and test an SVM model. The created model will be evaluated on the test set using evaluation accuracy, precision, recall, and F1 score to determine the performance of the SVM model developed to detect cyberbullying. Results: The results showed that SVM is a suitable technique for sentiment analysis of cyberbullying tweets. The model had an accuracy of 82.3% in detecting cyberbullying, with a precision of 0.82, recall of 0.82, and F1-score of 0.83. Conclusions: The study demonstrates the feasibility of using SVM for sentimental analysis of cyberbullying tweets. The high accuracy of the SVM model suggests that it can be used to build automated systems for detecting cyberbullying. The findings highlight the importance of developing tools to detect and address cyberbullying in the online world. The use of sentimental analysis and SVM has the potential to make a significant contribution to the fight against cyberbullying.

Object detection is a fundamental task in computer vision, with significant applications in various domains. However, the reliance on large-scale annotated data and computational resource demands poses challenges in practical implementation. This thesis aims to address these complexities by exploring self-supervised training approaches for the detection transformer(DETR) family of object detectors. The project investigates the necessity of training the backbone under a semi-supervised setting and explores the benefits of initializing scene graph generation architectures with pretrained DETReg and DETR models for faster training convergence and reduced computational resource requirements. The significance of this research lies in the potential to mitigate the dependence on annotated data and make deep learning techniques more accessible to researchers and practitioners. By overcoming the limitations of data and computational resources, this thesis contributes to the accessibility of DETR and encourages a more sustainable and inclusive approach to deep learning research. / Objektigenkänning är en grundläggande uppgift inom datorseende, med betydande tillämpningar inom olika domäner. Dock skapar beroendet av storskaliga annoterade data och krav på datorkraft utmaningar i praktisk implementering. Denna avhandling syftar till att ta itu med dessa komplexiteter genom att utforska självövervakade utbildningsmetoder för detektions transformer (DETR) familjen av objektdetektorer. Projektet undersöker nödvändigheten av att träna ryggraden under en semi-övervakad inställning och utforskar fördelarna med att initiera scenegrafgenereringsarkitekturer med förtränade DETReg-modeller för snabbare konvergens av träning och minskade krav på datorkraft. Betydelsen av denna forskning ligger i potentialen att mildra beroendet av annoterade data och göra djupinlärningstekniker mer tillgängliga för forskare och utövare. Genom att övervinna begränsningarna av data och datorkraft, bidrar denna avhandling till tillgängligheten av DETR och uppmuntrar till en mer hållbar och inkluderande inställning till djupinlärning forskning.

Product matching in e-commerce is an area that faces more and more challenges with growth in the e-commerce marketplace as well as variation in the quality of data available online for each product. Product matching for e-commerce provides competitive possibilities for vendors and flexibility for customers by identifying identical products from different sources. Traditional methods in product matching are often conducted through rule-based methods and methods tackling the issue through machine learning usually do so through unimodal systems. Moreover, existing methods would tackle the issue through product identifiers which are not always unified for each product. This thesis provides multimodal approaches through product name, description, and image to the problem area of product matching that outperforms unimodal approaches. Three multimodal approaches were taken, one unsupervised and two supervised. The unsupervised approach uses straight-forward embedding space to nearest neighbor search that provides better results than unimodal approaches. One of the supervised multimodal approaches uses Siamese network on the embedding space which outperforms the unsupervised multi- modal approach. Finally, the last supervised approach instead tackles the issue by exploiting distance differences in each modality through logistic regression and a decision system that provided the best results. / Produktmatchning inom e-handel är ett område som möter fler och fler utmaningar med hänsyn till den tillväxt som e-handelsmarknaden undergått och fortfarande undergår samt variation i kvaliteten på den data som finns tillgänglig online för varje produkt. Produktmatchning inom e-handel är ett område som ger konkurrenskraftiga möjligheter för leverantörer och flexibilitet för kunder genom att identifiera identiska produkter från olika källor. Traditionella metoder för produktmatchning genomfördes oftast genom regelbaserade metoder och metoder som utnyttjar maskininlärning gör det vanligtvis genom unimodala system. Dessutom utnyttjar mestadels av befintliga metoder produktidentifierare som inte alltid är enhetliga för varje produkt mellan olika källor. Denna studie ger istället förslag till multimodala tillvägagångssätt som istället använder sig av produktnamn, produktbeskrivning och produktbild för produktmatchnings-problem vilket ger bättre resultat än unimodala metoder. Tre multimodala tillvägagångssätt togs, en unsupervised och två supervised. Den unsupervised metoden använder embeddings vektorerna rakt av för att göra en nearest neighborsökning vilket gav bättre resultat än unimodala tillvägagångssätt. Ena supervised multimodal tillvägagångssätten använder siamesiska nätverk på embedding utrymmet vilket gav resultat som överträffade den unsupervised multimodala tillvägagångssättet. Slutligen tar den sista supervised metoden istället avståndsskillnader i varje modalitet genom logistisk regression och ett beslutssystem som gav bästa resultaten.

No description available.