scholarly journals Zero-Shot Human Activity Recognition Using Non-Visual Sensors

Sensors ◽  
2020 ◽  
Vol 20 (3) ◽  
pp. 825 ◽  
Author(s):  
Fadi Al Machot ◽  
Mohammed R. Elkobaisi ◽  
Kyandoghere Kyamakya
Keyword(s):  
Activity Recognition ◽  
High Performance ◽  
Real Life ◽  
Machine Learning Algorithms ◽  
Training Data ◽  
Sensor Data ◽  
Sensor Technology ◽  
Training Dataset ◽  
Data Sets ◽  
Data Set

Due to significant advances in sensor technology, studies towards activity recognition have gained interest and maturity in the last few years. Existing machine learning algorithms have demonstrated promising results by classifying activities whose instances have been already seen during training. Activity recognition methods based on real-life settings should cover a growing number of activities in various domains, whereby a significant part of instances will not be present in the training data set. However, to cover all possible activities in advance is a complex and expensive task. Concretely, we need a method that can extend the learning model to detect unseen activities without prior knowledge regarding sensor readings about those previously unseen activities. In this paper, we introduce an approach to leverage sensor data in discovering new unseen activities which were not present in the training set. We show that sensor readings can lead to promising results for zero-shot learning, whereby the necessary knowledge can be transferred from seen to unseen activities by using semantic similarity. The evaluation conducted on two data sets extracted from the well-known CASAS datasets show that the proposed zero-shot learning approach achieves a high performance in recognizing unseen (i.e., not present in the training dataset) new activities.

scholarly journals DOLARS, a Distributed On-Line Activity Recognition System by Means of Heterogeneous Sensors in Real-Life Deployments—A Case Study in the Smart Lab of The University of Almería

Sensors ◽  
2021 ◽  
Vol 21 (2) ◽  
pp. 405
Author(s):  
Marcos Lupión ◽  
Javier Medina-Quero ◽  
Juan F. Sanjuan ◽  
Pilar M. Ortigosa
Keyword(s):  
Real Time ◽  
Activity Recognition ◽  
Real Life ◽  
Recognition System ◽  
Machine Learning Algorithms ◽  
Sensor Data ◽  
Heterogeneous Sensors ◽  
On Line ◽  
The University

Activity Recognition (AR) is an active research topic focused on detecting human actions and behaviours in smart environments. In this work, we present the on-line activity recognition platform DOLARS (Distributed On-line Activity Recognition System) where data from heterogeneous sensors are evaluated in real time, including binary, wearable and location sensors. Different descriptors and metrics from the heterogeneous sensor data are integrated in a common feature vector whose extraction is developed by a sliding window approach under real-time conditions. DOLARS provides a distributed architecture where: (i) stages for processing data in AR are deployed in distributed nodes, (ii) temporal cache modules compute metrics which aggregate sensor data for computing feature vectors in an efficient way; (iii) publish-subscribe models are integrated both to spread data from sensors and orchestrate the nodes (communication and replication) for computing AR and (iv) machine learning algorithms are used to classify and recognize the activities. A successful case study of daily activities recognition developed in the Smart Lab of The University of Almería (UAL) is presented in this paper. Results present an encouraging performance in recognition of sequences of activities and show the need for distributed architectures to achieve real time recognition.

scholarly journals Multimodal Sensor Data Fusion for Activity Recognition Using Filtered Classifier

Proceedings ◽  
2018 ◽  
Vol 2 (19) ◽  
pp. 1262 ◽  
Author(s):  
Muhammad Razzaq ◽  
Ian Cleland ◽  
Chris Nugent ◽  
Sungyoung Lee
Keyword(s):  
Activity Recognition ◽  
Confusion Matrix ◽  
Machine Learning Algorithms ◽  
Training Data ◽  
Sensor Data ◽  
Test Dataset ◽  
Sensory Data ◽  
Multimodal Sensors ◽  
Imbalanced Class Distribution ◽  
Single Modality

Activity recognition (AR) is a subtask in pervasive computing and context-aware systems, which presents the physical state of human in real-time. These systems offer a new dimension to the widely spread applications by fusing recognized activities obtained from the raw sensory data generated by the obtrusive as well as unobtrusive revolutionary digital technologies. In recent years, an exponential growth has been observed for AR technologies and much literature exists focusing on applying machine learning algorithms on obtrusive single modality sensor devices. However, University of Jaén Ambient Intelligence (UJAmI), a Smart Lab in Spain has initiated a 1st UCAmI Cup challenge by sharing aforementioned varieties of the sensory data in order to recognize the human activities in the smart environment. This paper presents the fusion, both at the feature level and decision level for multimodal sensors by preprocessing and predicting the activities within the context of training and test datasets. Though it achieves 94% accuracy for training data and 47% accuracy for test data. However, this study further evaluates post-confusion matrix also and draws a conclusion for various discrepancies such as imbalanced class distribution within the training and test dataset. Additionally, this study also highlights challenges associated with the datasets for which, could improve further analysis.

scholarly journals A new concordant partial AUC and partial c statistic for imbalanced data in the evaluation of machine learning algorithms

2020 ◽  
Vol 20 (1) ◽  
Author(s):  
André M. Carrington ◽  
Paul W. Fieguth ◽  
Hammad Qazi ◽  
Andreas Holzinger ◽  
Helen H. Chen ◽  
...  
Keyword(s):  
Roc Curve ◽  
Diagnostic Testing ◽  
Real Life ◽  
Imbalanced Data ◽  
Machine Learning Algorithms ◽  
Data Sets ◽  
Data Set ◽  
Partial Auc ◽  
C Statistic ◽  
Future Work

Abstract Background In classification and diagnostic testing, the receiver-operator characteristic (ROC) plot and the area under the ROC curve (AUC) describe how an adjustable threshold causes changes in two types of error: false positives and false negatives. Only part of the ROC curve and AUC are informative however when they are used with imbalanced data. Hence, alternatives to the AUC have been proposed, such as the partial AUC and the area under the precision-recall curve. However, these alternatives cannot be as fully interpreted as the AUC, in part because they ignore some information about actual negatives. Methods We derive and propose a new concordant partial AUC and a new partial c statistic for ROC data—as foundational measures and methods to help understand and explain parts of the ROC plot and AUC. Our partial measures are continuous and discrete versions of the same measure, are derived from the AUC and c statistic respectively, are validated as equal to each other, and validated as equal in summation to whole measures where expected. Our partial measures are tested for validity on a classic ROC example from Fawcett, a variation thereof, and two real-life benchmark data sets in breast cancer: the Wisconsin and Ljubljana data sets. Interpretation of an example is then provided. Results Results show the expected equalities between our new partial measures and the existing whole measures. The example interpretation illustrates the need for our newly derived partial measures. Conclusions The concordant partial area under the ROC curve was proposed and unlike previous partial measure alternatives, it maintains the characteristics of the AUC. The first partial c statistic for ROC plots was also proposed as an unbiased interpretation for part of an ROC curve. The expected equalities among and between our newly derived partial measures and their existing full measure counterparts are confirmed. These measures may be used with any data set but this paper focuses on imbalanced data with low prevalence. Future work Future work with our proposed measures may: demonstrate their value for imbalanced data with high prevalence, compare them to other measures not based on areas; and combine them with other ROC measures and techniques.

scholarly journals Open-Source Data Collection and Data Sets for Activity Recognition in Smart Homes

Sensors ◽  
2020 ◽  
Vol 20 (3) ◽  
pp. 879 ◽  
Author(s):  
Uwe Köckemann ◽  
Marjan Alirezaie ◽  
Jennifer Renoux ◽  
Nicolas Tsiftes ◽  
Mobyen Uddin Ahmed ◽  
...  
Keyword(s):  
Data Collection ◽  
Activity Recognition ◽  
Care Home ◽  
Open Data ◽  
Ground Truth ◽  
Smart Homes ◽  
Sensor Data ◽  
Data Sets ◽  
Data Set ◽  
Home Setting

As research in smart homes and activity recognition is increasing, it is of ever increasing importance to have benchmarks systems and data upon which researchers can compare methods. While synthetic data can be useful for certain method developments, real data sets that are open and shared are equally as important. This paper presents the E-care@home system, its installation in a real home setting, and a series of data sets that were collected using the E-care@home system. Our first contribution, the E-care@home system, is a collection of software modules for data collection, labeling, and various reasoning tasks such as activity recognition, person counting, and configuration planning. It supports a heterogeneous set of sensors that can be extended easily and connects collected sensor data to higher-level Artificial Intelligence (AI) reasoning modules. Our second contribution is a series of open data sets which can be used to recognize activities of daily living. In addition to these data sets, we describe the technical infrastructure that we have developed to collect the data and the physical environment. Each data set is annotated with ground-truth information, making it relevant for researchers interested in benchmarking different algorithms for activity recognition.

Liquid Pipeline Rupture Detection Using Multiple Artificial Intelligence Classifiers During Steady-State and Transient Operations

2020 ◽  
Author(s):  
Christopher MacDonald ◽  
Michael Yang ◽  
Shawn Learn ◽  
Ron Hugo ◽  
Simon Park
Keyword(s):  
Artificial Intelligence ◽  
Training Data ◽  
Sensor Data ◽  
Data Sets ◽  
Data Set ◽  
Transient Model ◽  
Rule Based ◽  
Detection Systems ◽  
Pump Station ◽  
Neuro Fuzzy

Abstract There are several challenges associated with existing rupture detection systems such as their inability to accurately detect during transient (such as pump dynamics) conditions, delayed responses and their inability to transfer models to different pipeline configurations easily. To address these challenges, we employ multiple Artificial Intelligence (AI) classifiers that rely on pattern recognitions instead of traditional operator-set thresholds. AI techniques, consisting of two-dimensional (2D) Convolutional Neural Networks (CNN) and Adaptive Neuro Fuzzy Interface Systems (ANFIS), are used to mimic processes performed by operators during a rupture event. This includes both visualization (using CNN) and rule-based decision making (using ANFIS). The system provides a level of reasoning to an operator through the use of the rule-based AI system. Pump station sensor data is non-dimensionalized prior to AI processing, enabling application to pipeline configurations outside of the training data set. AI algorithms undergo testing and training using two data sets: laboratory-collected data that mimics transient pump-station operations and real operator data that includes Real Time Transient Model (RTTM) simulated ruptures. The use of non-dimensional sensor data enables the system to detect ruptures from pipeline data not used in the training process.

scholarly journals Multiclass Sentiment Analysis of Social Media Data using Neural Networks

2019 ◽  
Vol 9 (1S3) ◽  
pp. 57-62
Keyword(s):  
Neural Network ◽  
Sentiment Analysis ◽  
Social Networking Sites ◽  
Opinion Mining ◽  
Binary Classification ◽  
Multiclass Classification ◽  
Machine Learning Algorithms ◽  
Training Data ◽  
Training Dataset ◽  
Data Sets

Sentiment analysis, also known as Opinion Mining is one of the hottest topic Nowadays. in various social networking sites is one of the hottest topic and field nowadays. Here, we are using Twitter, the biggest web destinations for people to communicate with each other to perform the sentiment analysis and opinion mining by extracting the tweets by various users. The users can post brief text updates in twitter as it only allows 140 characters in one text message. Hashtags helps to search for tweets dealing with the specified subject. In previous researches, binary classification usually relies on the sentiment polarity(Positive , Negative and Neutral). The advantage is that multiple meaning of the same world might have different polarity, so it can be easily identified. In Multiclass classification, many tweets of one class are classified as if they belong to the others. The Neutral class presented the lowest precision in all the researches happened in this particular area. The set of tweets containing text and emoticon data will be classified into 13 classes. From each tweet, we extract different set of features using one hot encoding algorithm and use machine learning algorithms to perform classification. The entire tweets will be divided into training data sets and testing data sets. Training dataset will be pre-processed and classified using various Artificial Neural Network algorithms such as Reccurent Neural Network, Convolutional Neural Network etc. Moreover, the same procedure will be followed for the Text and Emoticon data. The developed model or system will be tested using the testing dataset. More precise and correct accuracy can be obtained or experienced using this multiclass classification of text and emoticons. 4 Key performance indicators will be used to evaluate the effectiveness of the corresponding approach.

scholarly journals Translating Videos into Synthetic Training Data for Wearable Sensor-Based Activity Recognition Systems Using Residual Deep Convolutional Networks

2021 ◽  
Vol 11 (7) ◽  
pp. 3094
Author(s):  
Vitor Fortes Rey ◽  
Kamalveer Kaur Garewal ◽  
Paul Lukowicz
Keyword(s):  
Computer Vision ◽  
Regression Model ◽  
Activity Recognition ◽  
Language Processing ◽  
Large Scale ◽  
Simulated Data ◽  
Training Data ◽  
Sensor Data ◽  
Activity Data ◽  
Data Set

Human activity recognition (HAR) using wearable sensors has benefited much less from recent advances in Deep Learning than fields such as computer vision and natural language processing. This is, to a large extent, due to the lack of large scale (as compared to computer vision) repositories of labeled training data for sensor-based HAR tasks. Thus, for example, ImageNet has images for around 100,000 categories (based on WordNet) with on average 1000 images per category (therefore up to 100,000,000 samples). The Kinetics-700 video activity data set has 650,000 video clips covering 700 different human activities (in total over 1800 h). By contrast, the total length of all sensor-based HAR data sets in the popular UCI machine learning repository is less than 63 h, with around 38 of those consisting of simple mode of locomotion activities like walking, standing or cycling. In our research we aim to facilitate the use of online videos, which exist in ample quantities for most activities and are much easier to label than sensor data, to simulate labeled wearable motion sensor data. In previous work we already demonstrated some preliminary results in this direction, focusing on very simple, activity specific simulation models and a single sensor modality (acceleration norm). In this paper, we show how we can train a regression model on generic motions for both accelerometer and gyro signals and then apply it to videos of the target activities to generate synthetic Inertial Measurement Units (IMU) data (acceleration and gyro norms) that can be used to train and/or improve HAR models. We demonstrate that systems trained on simulated data generated by our regression model can come to within around 10% of the mean F1 score of a system trained on real sensor data. Furthermore, we show that by either including a small amount of real sensor data for model calibration or simply leveraging the fact that (in general) we can easily generate much more simulated data from video than we can collect its real version, the advantage of the latter can eventually be equalized.

Transferable Pipeline Rupture Detection Using Multiple Artificial Intelligence Classifiers During Transient Operations

2021 ◽  
Author(s):  
Christopher Macdonald ◽  
Jaehyun Yang ◽  
Shawn Learn ◽  
Simon S. Park ◽  
Ronald J. Hugo
Keyword(s):  
Artificial Intelligence ◽  
Response Times ◽  
Training Data ◽  
Sensor Data ◽  
Data Sets ◽  
Flow Loop ◽  
Data Set ◽  
Transient Model ◽  
Rule Based ◽  
Pump Station

Abstract There are several challenges associated with existing pipeline rupture detection systems, including an inability to accurately detect during transient conditions (such as changes in pump operating points), an inability to easily transfer from one pipeline configuration to another, and relatively slow response times. To address these challenges, we employ multiple Artificial Intelligence (AI) classifiers that rely on pattern recognition instead of traditional operator-set thresholds. AI techniques, consisting of two-dimensional (2D) Convolutional Neural Networks (CNN) and Adaptive Neuro Fuzzy Interface Systems (ANFIS), are used to mimic processes performed by operators during a rupture event. This includes both visualization (using CNN) and rule-based decision making (using ANFIS). The system provides a level of reasoning to an operator through the use of rule-based AI. Pump station sensor data is non-dimensionalized prior to AI processing, enabling pipeline configurations outside of the training data set, independent of geometry, length, and medium. AI algorithms undergo testing and training using two data sets: laboratory-collected flow loop data that mimics transient pump-station operations and real operator data that include simulated ruptures using the Real Time Transient Model (RTTM). The multiple AI classifier results are fused together to provide higher reliability especially detecting ruptures from pipeline data not used in the training process.

Practical strategies for classification of unexploded ordnance

Geophysics ◽  
2013 ◽  
Vol 78 (1) ◽  
pp. E41-E46 ◽  
Author(s):  
Laurens Beran ◽  
Barry Zelt ◽  
Leonard Pasion ◽  
Stephen Billings ◽  
Kevin Kingdon ◽  
...  
Keyword(s):  
Classification Performance ◽  
Training Data ◽  
Sensor Data ◽  
Data Sets ◽  
Unexploded Ordnance ◽  
Model Quality ◽  
Data Set ◽  
Time Domain Electromagnetic

We have developed practical strategies for discriminating between buried unexploded ordnance (UXO) and metallic clutter. These methods are applicable to time-domain electromagnetic data acquired with multistatic, multicomponent sensors designed for UXO classification. Each detected target is characterized by dipole polarizabilities estimated via inversion of the observed sensor data. The polarizabilities are intrinsic target features and so are used to distinguish between UXO and clutter. We tested this processing with four data sets from recent field demonstrations, with each data set characterized by metrics of data and model quality. We then developed techniques for building a representative training data set and determined how the variable quality of estimated features affects overall classification performance. Finally, we devised a technique to optimize classification performance by adapting features during target prioritization.

scholarly journals Contrastive Predictive Coding for Human Activity Recognition

2021 ◽  
Vol 5 (2) ◽  
pp. 1-26
Author(s):  
Harish Haresamudram ◽  
Irfan Essa ◽  
Thomas Plötz
Keyword(s):  
Ubiquitous Computing ◽  
Activity Recognition ◽  
Human Activity ◽  
Recognition Performance ◽  
Temporal Structure ◽  
Predictive Coding ◽  
Real Life ◽  
Human Activity Recognition ◽  
Training Data ◽  
Sensor Data

Feature extraction is crucial for human activity recognition (HAR) using body-worn movement sensors. Recently, learned representations have been used successfully, offering promising alternatives to manually engineered features. Our work focuses on effective use of small amounts of labeled data and the opportunistic exploitation of unlabeled data that are straightforward to collect in mobile and ubiquitous computing scenarios. We hypothesize and demonstrate that explicitly considering the temporality of sensor data at representation level plays an important role for effective HAR in challenging scenarios. We introduce the Contrastive Predictive Coding (CPC) framework to human activity recognition, which captures the temporal structure of sensor data streams. Through a range of experimental evaluations on real-life recognition tasks, we demonstrate its effectiveness for improved HAR. CPC-based pre-training is self-supervised, and the resulting learned representations can be integrated into standard activity chains. It leads to significantly improved recognition performance when only small amounts of labeled training data are available, thereby demonstrating the practical value of our approach. Through a series of experiments, we also develop guidelines to help practitioners adapt and modify the framework towards other mobile and ubiquitous computing scenarios.

Sign in / Sign up

Export Citation Format

Share Document