Indiana University-Purdue University Indianapolis (IUPUI) / For pedestrians and autonomous vehicles (AVs) to co-exist harmoniously and safely in the real-world, AVs will need to not only react to pedestrian actions, but also anticipate their intentions. In this thesis, we propose to use rich visual and pedestrian-environment interaction features to improve pedestrian crossing intention prediction from the ego-view.We do so by combining visual feature extraction, graph modeling of scene objects and their relationships, and feature encoding as comprehensive inputs for an LSTM encoder-decoder network. Pedestrians react and make decisions based on their surrounding environment, and the behaviors of other road users around them. The human-human social relationship has al-ready been explored for pedestrian trajectory prediction from the bird’s eye view in stationary cameras. However, context and pedestrian-environment relationships are often missing incurrent research into pedestrian trajectory, and intention prediction from the ego-view. To map the pedestrian’s relationship to its surrounding objects we use a star graph with the pedestrian in the center connected to all other road objects/agents in the scene. The pedestrian and road objects/agents are represented in the graph through visual features extracted using state of the art deep learning algorithms. We use graph convolutional networks, and graph autoencoders to encode the star graphs in a lower dimension. Using the graph en-codings, pedestrian bounding boxes, and human pose estimation, we propose a novel model that predicts pedestrian crossing intention using not only the pedestrian’s action behaviors(bounding box and pose estimation), but also their relationship to their environment. Through tuning hyperparameters, and experimenting with different graph convolutions for our graph autoencoder, we are able to improve on the state of the art results. Our context-driven method is able to outperform current state of the art results on benchmark datasetPedestrian Intention Estimation (PIE). The state of the art is able to predict pedestrian crossing intention with a balanced accuracy (to account for dataset imbalance) score of 0.61, while our best performing model has a balanced accuracy score of 0.79. Our model especially outperforms in no crossing intention scenarios with an F1 score of 0.56 compared to the state of the art’s score of 0.36. Additionally, we also experiment with training the state of the art model and our model to predict pedestrian crossing action, and intention jointly. While jointly predicting crossing action does not help improve crossing intention prediction, it is an important distinction to make between predicting crossing action versus intention.

No description available.

No description available.

No description available.

Scalp electroencephalograms (EEGs) are used in a broad range of health care institutions to monitor and record electrical activity in the brain. EEGs are essential in diagnosis of clinical conditions such as epilepsy, seizure, coma, encephalopathy, and brain death. Manual scanning and interpretation of EEGs is time-consuming since these recordings may last hours or days. It is also an expensive process as it requires highly trained experts. Therefore, high performance automated analysis of EEGs can reduce time to diagnosis and enhance real-time applications by identifying sections of the signal that need further review.Automatic analysis of clinical EEGs is a very difficult machine learning problem due to the low fidelity of a scalp EEG signal. Commercially available automated seizure detection systems suffer from unacceptably high false alarm rates. Many signal processing methods have been developed over the years including time-frequency processing, wavelet analysis and autoregressive spectral analysis. Though there has been significant progress in machine learning technology in recent years, use of automated technology in clinical settings is limited, mainly due to unacceptably high false alarm rates. Further, state of the art machine learning algorithms that employ high dimensional models have not previously been utilized in EEG analysis because there has been a lack of large databases that accurately characterize clinical operating conditions. Deep learning approaches can be viewed as a broad family of neural network algorithms that use many layers of nonlinear processing units to learn a mapping between inputs and outputs. Deep learning-based systems have generated significant improvements in performance for sequence recognitions tasks for temporal signals such as speech and for image analysis applications that can exploit spatial correlations, and for which large amounts of training data exists. The primary goal of our proposed research is to develop deep learning-based architectures that capture spatial and temporal correlations in an EEG signal. We apply these architectures to the problem of automated seizure detection for adult EEGs. The main contribution of this work is the development of a high-performance automated EEG analysis system based on principles of machine learning and big data that approaches levels of performance required for clinical acceptance of the technology. In this work, we explore a combination of deep learning-based architectures. First, we present a hybrid architecture that integrates hidden Markov models (HMMs) for sequential decoding of EEG events with a deep learning-based postprocessing that incorporates temporal and spatial context. This system automatically processes EEG records and classifies three patterns of clinical interest in brain activity that might be useful in diagnosing brain disorders: spike and/or sharp waves, generalized periodic epileptiform discharges and periodic lateralized epileptiform discharges. It also classifies three patterns used to model the background EEG activity: eye movement, artifacts, and background. Our approach delivers a sensitivity above 90% while maintaining a specificity above 95%. Next, we replace the HMM component of the system with a deep learning architecture that exploits spatial and temporal context. We study how effectively these architectures can model context. We introduce several architectures including a novel hybrid system that integrates convolutional neural networks with recurrent neural networks to model both spatial relationships (e.g., cross-channel dependencies) and temporal dynamics (e.g., spikes). We also propose a topology-preserving architecture for spatio-temporal sequence recognition that uses raw data directly rather than low-level features. We show this model learns representations directly from raw EEGs data and does not need to use predefined features. In this study, we use the Temple University EEG (TUEG) Corpus, supplemented with data from Duke University and Emory University, to evaluate the performance of these hybrid deep structures. We demonstrate that performance of a system trained only on Temple University Seizure Corpus (TUSZ) data transfers to a blind evaluation set consisting of the Duke University Seizure Corpus (DUSZ) and the Emory University Seizure Corpus (EUSZ). This type of generalization is very important since complex high-dimensional deep learning systems tend to overtrain. We also investigate the robustness of this system to mismatched conditions (e.g., train on TUSZ, evaluate on EUSZ). We train a model on one of three available datasets and evaluate the trained model on the other two datasets. These datasets are recorded from different hospitals, using a variety of devices and electrodes, under different circumstances and annotated by different neurologists and experts. Therefore, these experiments help us to evaluate the impact of the dataset on our training process and validate our manual annotation process. Further, we introduce methods to improve generalization and robustness. We analyze performance to gain additional insight into what aspects of the signal are being modeled adequately and where the models fail. The best results for automatic seizure detection achieved in this study are 45.59% with 12.24 FA per 24 hours on TUSZ, 45.91% with 11.86 FAs on DUSZ, and 62.56% with 11.26 FAs on EUSZ. We demonstrate that the performance of the deep recurrent convolutional structure presented in this study is statistically comparable to the human performance on the same dataset. / Electrical and Computer Engineering

<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.

The digitalization of the energy industry has made tremendous energy data available. This data is utilized across the entire energy value chain to provide value for customers and energy providers. One area that has gained recent attention in the energy industry is the electricity load forecasting for better scheduling and bidding on the electricity market. However, the electricity data that is used for forecasting is prone to have anomalies, which can affect the accuracy of forecasts. In this thesis we propose two anomaly detection methods to tackle the issue of anomalies in electricity demand data. We propose Long short-term memory (LSTM) and Feed-forward neural network (FFNN) based methods, and compare their anomaly detection performance on two real-world electricity demand datasets. Our results indicate that the LSTM model tends to produce a more robust behavior than the FFNN model on the dataset with regular daily and weekly patterns. However, there was no significant difference between the performance of the two models when the data was noisy and showed no regular patterns. While our results suggest that the LSTM model is effective when a regular pattern in data is present, the results were not found to be statistically significant to claim superiority of LSTM over FFNN. / Digitaliseringen inom energibranschen har tillgängliggjort enorma mängder energidata. Dessa data används över hela värdekedjan för energisystem i syfte att skapa värde för kunder och energileverantörer. Ett område som nyligen uppmärksammats inom energibranschen är att skapa prognoser för elbelastning för bättre schemaläggning och budgivning på elmarknaden. Data som används för sådana prognoser är dock benägna att ha avvikelser, vilket kan påverka prognosernas noggrannhet. I det här examensarbetet föreslår vi två metoder för detektering av avvikelser för att ta itu med frågan om avvikelser i data för elektricitetsbehov. Vi föreslår metoder baserade på Long short-term memory (LSTM) och Feedforward neural network (FFNN) och jämför dess prestanda att upptäcka avvikelser på två verkliga databanker över elbehovsdata. Våra resultat indikerar att LSTM-modellen tenderar att producera ett mer robust beteende än FFNN-modellen på data med upprepande dagliga samt veckovisa mönster. Det fanns dock ingen signifikant skillnad mellan prestanda för de två modellerna när data inte uppvisade regelbunda mönster. Även om våra resultat antyder att LSTM-modellen är effektiv när ett regelbundet datamönster finns närvarande, var resultaten inte statistiskt signifikanta för att påstå överlägsenhet av LSTM jämfört med FFNN.

Customer churn is problematic for any business trying to expand their customer base. The acquisition of new customers to replace churned ones are associated with additional costs, whereas taking measures to retain existing customers may prove more cost efficient. As such, it is of interest to estimate the time until the occurrence of a potential churn for every customer in order to take preventive measures. The application of deep learning and machine learning to this type of problem using time series data is relatively new and there is a lot of recent research on this topic. This thesis is based on the assumption that early signs of churn can be detected by the temporal changes in customer behavior. Recurrent neural networks and more specifically long short-term memory (LSTM) and gated recurrent unit (GRU) are suitable contenders since they are designed to take the sequential time aspect of the data into account. Random forest (RF) and stochastic vector machine (SVM) are machine learning models that are frequently used in related research. The problem is solved through a classification approach, and a comparison is done with implementations using LSTM, GRU, RF, and SVM. According to the results, LSTM and GRU perform similarly while being slightly better than RF and SVM in the task of predicting customers that will churn in the coming six months, and that all models could potentially lead to cost savings according to simulations (using non-official but reasonable costs assigned to each prediction outcome). Predicting the time until churn is a more difficult problem and none of the models can give reliable estimates, but all models are significantly better than random predictions. / Kundbortfall är problematiskt för företag som försöker expandera sin kundbas. Förvärvandet av nya kunder för att ersätta förlorade kunder är associerat med extra kostnader, medan vidtagandet av åtgärder för att behålla kunder kan visa sig mer lönsamt. Som så är det av intresse att för varje kund ha pålitliga tidsestimat till en potentiell uppsägning kan tänkas inträffa så att förebyggande åtgärder kan vidtas. Applicerandet av djupinlärning och maskininlärning på denna typ av problem som involverar tidsseriedata är relativt nytt och det finns mycket ny forskning kring ämnet. Denna uppsats är baserad på antagandet att tidiga tecken på kundbortfall kan upptäckas genom kunders användarmönster över tid. Reccurent neural networks och mer specifikt long short-term memory (LSTM) och gated recurrent unit (GRU) är lämpliga modellval eftersom de är designade att ta hänsyn till den sekventiella tidsaspekten i tidsseriedata. Random forest (RF) och stochastic vector machine (SVM) är maskininlärningsmodeller som ofta används i relaterad forskning. Problemet löses genom en klassificeringsapproach, och en jämförelse utförs med implementationer av LSTM, GRU, RF och SVM. Resultaten visar att LSTM och GRU presterar likvärdigt samtidigt som de presterar bättre än RF och SVM på problemet om att förutspå kunder som kommer att säga upp sig inom det kommande halvåret, och att samtliga modeller potentiellt kan leda till kostnadsbesparingar enligt simuleringar (som använder icke-officiella men rimliga kostnader associerat till varje utfall). Att förutspå tid till en kunduppsägning är ett svårare problem och ingen av de framtagna modellerna kan ge pålitliga tidsestimat, men alla är signifikant bättre än slumpvisa gissningar.

The aim of this work is to develop efficient named entity recognition from the given text that in turn improves the performance of the systems that use natural language processing (NLP). The performance of IoT-based devices such as Alexa and Cortana significantly depends upon an efficient NLP model. To increase the capability of the smart IoT devices in comprehending the natural language, named entity recognition (NER) tools play an important role in these devices. In general, the NER is a two-step process that initially the proper nouns are identified from text and then classify them into predefined categories of entities such as person, location, measure, organization and time. NER is often performed as a subtask while processing natural languages which increases the accuracy level of a NLP task. In this paper, we propose deep neural network architecture for named entity recognition for the resource-scarce language Hindi, based on convolutional neural network (CNN), bidirectional long short-term memory (Bi-LSTM) neural network and conditional random field (CRF). In the proposed approach, initially, we use skip-gram word2vec model and GloVe model to represent words in semantic vectors which are further used in different deep neural network-based architectures. In the proposed approach, we use character- and word-level embedding to represent the text that includes information at fine-grained level. Due to the use of character-level embeddings, the proposed model is robust for the out-of-vocabulary words. Experimental results show that the combination of Bi-LSTM, CNN and CRF algorithms performs better as compared to the other baseline methods such as recurrent neural network, long short-term memory and Bi-LSTM individually.