Indiana University-Purdue University Indianapolis (IUPUI) / Action recognition has been an active research topic for over three decades. There are various applications of action recognition, such as surveillance, human-computer interaction, and content-based retrieval. Recently, research focuses on movies, web videos, and TV shows datasets. The nature of these datasets make action recognition very challenging due to scene variability and complexity, namely background clutter, occlusions, viewpoint changes, fast irregular motion, and large spatio-temporal search space (articulation configurations and motions). The use of local space and time image features shows promising results, avoiding the cumbersome and often inaccurate frame-by-frame segmentation (boundary estimation). We focus on two state of the art methods for the action classification problem: dense trajectories and recurrent neural networks (RNN). Dense trajectories use typical supervised training (e.g., with Support Vector Machines) of features such as 3D-SIFT, extended SURF, HOG3D, and local trinary patterns; the main idea is to densely sample these features in each frame and track them in the sequence based on optical flow. On the other hand, the deep neural network uses the input frames to detect action and produce part proposals, i.e., estimate information on body parts (shapes and locations). We compare qualitatively and numerically these two approaches, indicative to what is used today, and describe our conclusions with respect to accuracy and efficiency.

<p>The rapid expansion of intra-vehicle networks has increased the number of threats to such networks. Most modern vehicles implement various physical and data-link layer technologies. Vehicles are becoming increasingly autonomous and connected. Controller Area Network (CAN) is a serial bus system that is used to connect sensors and controllers (Electronic Control Units – ECUs) within a vehicle. ECUs vary widely in processing power, storage, memory, and connectivity. The goal of this research is to design, implement, and test an efficient and effective intrusion detection system for intra-vehicle CANs. Such a system must be capable of detecting intrusions in almost real-time with minimal resources. The research proposes a specific type of recursive neural network called Long Short-Term Memory (LSTM) to detect anomalies. It also proposes a decision engine that will use LSTM-classified anomalies to detect intrusions by using multiple contextual parameters. We have conducted multiple experiments on the optimal choice of various LSTM hyperparameters. We have tested our classification algorithm and our decision engine using data from real automobiles. We will present the results of our experiments and analyze our findings. After detailed evaluation of our intrusion detection system, we believe that we have designed a vehicle security solution that meets all the outlined requirements and goals.</p>

No description available.

A proof-of-concept indirect tire-pressure monitoring system is developed using neural net- works to identify the tire pressure of a vehicle tire. A quarter-car model was developed with Matlab and Simulink to generate simulated accelerometer output data. Simulation data are used to train and evaluate a recurrent neural network with long short-term memory blocks (RNN-LSTM) and a convolutional neural network (CNN) developed in Python with Tensorflow. Bayesian Optimization via SigOpt was used to optimize training and model parameters. The predictive accuracy and training speed of the two models with various parameters are compared. Finally, future work and improvements are discussed.

The majority of manufacturers provide their customers with volume discounts for placing repeat purchases or placing larger orders. In today's highly competitive market, the topic of how precisely a big number of products should be grouped together naturally emerges.\\In this context, three research questions that were directly relevant to the setting were formulated and their answers were provided. In order to achieve this goal, a number of experiments were carried out. In this particular instance, an algorithm was developed that determines the order combination that is mathematically superior to all others. In this context, an annual order cost saving of 1.33\% could be achieved based on the orders from the year 2021. This could be accomplished without the utilisation of heuristics for a limited number of products at most. In addition, a number of other heuristics have been devised for higher order combination sets. In addition, two other approaches to demand forecasting were investigated, and it was discovered that the time series in this particular instance was insufficient for the application of an RNN-LSTM model.

Text-to-speech (TTS) systems have entered our daily lives in the form of smart assistants and many other applications. Contemporary re- search applies machine learning and artificial neural networks (ANNs) to synthesize speech. It has been shown that these systems outperform the older concatenative and parametric methods. In this paper, ANN-based methods for speech synthesis are ex- plored and one of the methods is implemented for the Swedish lan- guage. The implemented method is dubbed “Tacotron” and is a first step towards end-to-end ANN-based TTS which puts many differ- ent ANN-techniques to work. The resulting system is compared to a parametric TTS through a strength-of-preference test that is carried out with 20 Swedish speaking subjects. A statistically significant pref- erence for the ANN-based TTS is found. Test subjects indicate that the ANN-based TTS performs better than the parametric TTS when it comes to audio quality and naturalness but sometimes lacks in intelli- gibility. / Talsynteser, också kallat TTS (text-to-speech) används i stor utsträckning inom smarta assistenter och många andra applikationer. Samtida forskning applicerar maskininlärning och artificiella neurala nätverk (ANN) för att utföra talsyntes. Det har visats i studier att dessa system presterar bättre än de äldre konkatenativa och parametriska metoderna. I den här rapporten utforskas ANN-baserade TTS-metoder och en av metoderna implementeras för det svenska språket. Den använda metoden kallas “Tacotron” och är ett första steg mot end-to-end TTS baserat på neurala nätverk. Metoden binder samman flertalet olika ANN-tekniker. Det resulterande systemet jämförs med en parametriskt TTS genom ett graderat preferens-test som innefattar 20 svensktalande försökspersoner. En statistiskt säkerställd preferens för det ANN- baserade TTS-systemet fastställs. Försökspersonerna indikerar att det ANN-baserade TTS-systemet presterar bättre än det parametriska när det kommer till ljudkvalitet och naturlighet men visar brister inom tydlighet.

The smartphone is becoming a common device carried by hundreds of millions of individual humans worldwide, and is used to accomplish a multitude of different tasks like basic communication, internet browsing, online shopping and fitness tracking. Limited by its small size and tight energy storage, the human-smartphone interface is largely bound to the smartphones small screens and simple keypads. This prohibits introducing new rich ways of interaction with smartphones.   The industry and research community are working extensively to find ways to enrich the human-smartphone interface by either seizing the existing smartphones resources like microphones, cameras and inertia sensors, or by introducing new specialized sensing capabilities into the smartphones like compact gesture sensing radar devices.   The prevalence of Radio Frequency (RF) signals and their limited power needs, led us towards investigating using RF signals received by smartphones to recognize gestures and activities around smartphones. This thesis introduces a solution for recognizing touch-less dynamic hand gestures from the Wi-Fi Received Signal Strength (RSS) received by the smartphone using a recurrent neural network (RNN) based probabilistic model. Unlike other Wi-Fi based gesture recognition solutions, the one introduced in this thesis does not require a change to the smartphone hardware or operating system, and performs the hand gesture recognition without interfering with the normal operation of other smartphone applications.   The developed hand gesture recognition solution achieved a mean accuracy of 78% detecting and classifying three hand gestures in an online setting involving different spatial and traffic scenarios between the smartphone and Wi-Fi access points (AP). Furthermore the characteristics of the developed solution were studied, and a set of improvements have been suggested for further future work. / Smarta telefoner bärs idag av hundratals miljoner människor runt om i världen, och används för att utföra en mängd olika uppgifter, så som grundläggande kommunikation, internetsökning och online-inköp. På grund av begränsningar i storlek och energilagring är människa-telefon-gränssnitten dock i hög grad begränsade till de förhållandevis små skärmarna och enkla knappsatser.   Industrin och forskarsamhället arbetar för att hitta vägar för att förbättra och bredda gränssnitten genom att antingen använda befintliga resurser såsom mikrofoner, kameror och tröghetssensorer, eller genom att införa nya specialiserade sensorer i telefonerna, som t.ex. kompakta radarenheter för gestigenkänning.   Det begränsade strömbehovet hos radiofrekvenssignaler (RF) inspirerade oss till att undersöka om dessa kunde användas för att känna igen gester och aktiviteter i närheten av telefoner. Denna rapport presenterar en lösning för att känna igen gester med hjälp av ett s.k. recurrent neural network (RNN). Till skillnad från andra Wi-Fi-baserade lösningar kräver denna lösning inte en förändring av vare sig hårvara eller operativsystem, och ingenkänningen genomförs utan att inverka på den normala driften av andra applikationer på telefonen.   Den utvecklade lösningen når en genomsnittlig noggranhet på 78% för detektering och klassificering av tre olika handgester, i ett antal olika konfigurationer vad gäller telefon och Wi-Fi-sändare. Rapporten innehåller även en analys av flera olika egenskaper hos den föreslagna lösningen, samt förslag till vidare arbete.

Churn prediction is used to identify customers that are becoming less loyal and is an important tool for companies that want to stay competitive in a rapidly growing market. In retail, a dynamic definition of churn is needed to identify churners correctly. Customer Lifetime Value (CLV) is the monetary value of a customer relationship. No change in CLV for a given customer indicates a decrease in loyalty. This thesis proposes a novel approach to churn prediction. The proposed model uses a Recurrent Neural Network to identify churners based on Customer Lifetime Value time series regression. The results show that the model performs better than random. This thesis also investigated the use of the K-means algorithm as a replacement to a rule-extraction algorithm. The K-means algorithm contributed to a more comprehensive analytical context regarding the churn prediction of the proposed model. / Illojalitet prediktering används för att identifiera kunder som är påväg att bli mindre lojala och är ett hjälpsamt verktyg för att ett företag ska kunna driva en konkurrenskraftig verksamhet. I detaljhandel behöves en dynamisk definition av illojalitet för att korrekt kunna identifera illojala kunder. Kundens livstidsvärde är ett mått på monetärt värde av en kundrelation. En avstannad förändring av detta värde indikerar en minskning av kundens lojalitet. Denna rapport föreslår en ny metod för att utföra illojalitet prediktering. Den föreslagna metoden består av ett återkommande neuralt nätverk som används för att identifiera illojalitet hos kunder genom att prediktera kunders livstidsvärde. Resultaten visar att den föreslagna modellen presterar bättre jämfört med slumpmässig metod. Rapporten undersöker också användningen av en k-medelvärdesalgoritm som ett substitut för en regelextraktionsalgoritm. K-medelsalgoritm bidrog till en mer omfattande analys av illojalitet predikteringen.

Conformité Européenne (CE) approval is a complex task for producers in Europe. The producers need to search for necessary standard documents and do the tests by themselves. CE-CHECK is a website which provides document searching service, and the company engineers want to use machine learning methods to analysis the documents and the results can improve the searching system. The first task is to construct an auto keyword extraction system to analysis the standard documents. This paper performed three different machine learning methods: Conditional Random Field (CRF), joint-layer Recurrent Neural Network (RNN), and double directional Long Short-Term Memory network (Bi-LSTM), for this task and tested their performances. CRF is a traditional probabilistic model which is widely used in sequential processing. RNN and LSTM are neural network models which show impressive performance on Natural Language processing in recent years. The result of the tests was that Bi-LSTM had the best performance: the keyword extraction recall was 76.97% while RNN was 72.99% and CRF was 70.18%. In conclusion, Bi-LSTM is the best model for this keyword extraction task, and the accuracy is high enough to provide a reliable result. The model also has good robustness that it have excellent performance on documents in different fields. Bi-LSTM model can analysis all documents in less than five minutes while manual works need months, so it saved both time and cost. The results can be used in searching system and further document analysis. / Att få Conformité Européenne (CE)-godkännande är en komplicerad process för producenter i Europa. Producenterna måste söka efter nödvändiga dokument för standarder samt utföra olika tester själva. CE-CHECK är en hemsida som erbjuder söktjänster för dokument. Företagets ingenjörer vill använda maskininlärningsmetoder för att analysera dokumenten då resultaten kan förbättra söksystemet. Den första uppgiften är att konstruera ett system som automatiskt extraherar nyckelord för att analysera dokument för standarder. Detta examensarbete använde tre olika maskininlärningsmetoder och testade deras prestanda: Conditional Random Field (CRF), joint-layer Recurrent Neural Network (RNN), samt Double directional Long Short-Term Memory network (Bi-LSTM). CRF är en traditionell probabilistisk modell som ofta används inom behandling av sekventiella data. RNN och LSTM är neurala nätverksmodeller som har visat imponerande resultat inom språkteknologi de senaste åren. Resultatet av undersökningen var att Bi-LSTM presterade bäst. Modellen lyckades extrahera 76.97% av nyckelorden medan resultatet för RNN var 72.99% och för CRF var det 70.18%. Slutsatsen blev således att Bi-LSTM är det bästa valet av modell för denna uppgift och dess exakthet är tillräckligt god för att producera pålitliga resultat. Modellen är även robust då den visar goda resultat på dokument från olika forskningsområden. Bi-LSTM kan analysera alla dokument på mindre än fem minuter medan manuellt arbete skulle kräva månader. Den minskar således både tidsåtgång och kostnad. Resultaten kan användas både i söksystem samt i vidare analys av dokument.

Background:The background of this research lies in detecting the images from satellites. The recognition of images from satellites has become increasingly importantdue to the vast amount of data that can be obtained from satellites. This thesisaims to develop a method for the recognition of images from satellites using machinelearning techniques. Objective:The main objective of this thesis is a unique approach to recognizingthe data with a CRNN algorithm that involves image recognition in satellite imagesusing machine learning, specifically the CRNN (Convolutional Recurrent Neural Network) architecture. The main task is classifying the images accurately, and this isachieved by utilizing object classification algorithms. The CRNN architecture ischosen because it can effectively extract features from satellite images using Convolutional Blocks and leverage the great memory power of the Long Short-TermMemory (LSTM) networks to connect the extracted features efficiently. The connected features improve the accuracy of our model significantly. Method:The proposed method involves doing a literature review to find currentimage recognition models and then experimentation by training a CRNN, CNN andRNN and then comparing their performance using metrics mentioned in the thesis work. Results:The performance of the proposed method is evaluated using various metrics, including precision, recall, F1 score and inference speed, on a large dataset oflabeled images. The results indicate that high accuracy is achieved in detecting andclassifying objects in satellite images through our approach. The potential utilization of our proposed method can span various applications such as environmentalmonitoring, urban planning, and disaster management. Conclusion:The classification on the satellite images is performed using the 2 datasetsfor ships and cars. The proposed architectures are CRNN, CNN, and RNN. These3 models are compared in order to find the best performing algorithm. The resultsindicate that CRNN has the best accuracy and precision and F1 score and inferencespeed, indicating a strong performance by the CRNN. Keywords: Comparison of CRNN, CNN, and RNN, Image recognition, MachineLearning, Algorithms,You Only Look Once. Version3, Satellite images, Aerial Images, Deep Learning