Sound event localization and detection based on deep learning

Acoustic source localization(ASL)and sound event detection(SED)are two widely pursued independent research fields.In recent years,in order to achieve a more complete spatial and temporal representation of sound field,sound event localization and detection(SELD)has become a very active research topic.This paper presents a deep learning-based multioverlapping sound event localization and detection algorithm in three-dimensional space.Log-Mel spectrum and generalized cross-correlation spectrum are joined together in channel dimension as input features.These features are classified and regressed in parallel after training by a neural network to obtain sound recognition and localization results respectively.The channel attention mechanism is also introduced in the network to selectively enhance the features containing essential information and suppress the useless features.Finally,a thourough comparison confirms the efficiency and effectiveness of the proposed SELD algorithm.Field experiments show that the proposed algorithm is robust to reverberation and environment and can achieve higher recognition and localization accuracy compared with the baseline method.

Suitable triggering algorithms for detecting strong ground motions using MEMS accelerometers

With the recent development of digital Micro Electro Mechanical System （MEMS） sensors, the cost of monitoring and detecting seismic events in real time can be greatly reduced. Ability of MEMS accelerograph to record a seismic event depends upon the efficiency of triggering algorithm, apart from the sensor＇s sensitivity. There are several classic triggering algorithms developed to detect seismic events, ranging from basic amplitude threshold to more sophisticated pattern recognition. Algorithms based on STA/LTA are reported to be computationally efficient for real time monitoring. In this paper, we analyzed several STA/LTA algorithms to check their efficiency and suitability using data obtained from the Quake Catcher Network （network of MEMS accelerometer stations）. We found that most of the STA/LTA algorithms are suitable for use with MEMS accelerometer data to accurately detect seismic events. However, the efficiency of any particular algorithm is found to be dependent on the parameter set used （i.e., window width of STA, LTA and threshold level）.

Word-Representation-Based Method for Extracting Organizational Events from Online Media

Online social media exhibit massive organizational event relevant messages, and the well categorized event information can be useful in many real-world applications. In this paper, we propose a research framework to extract high quality event information from massive online media data. The main contributions lie in two aspects: First, we present an event-extraction and event-categorization system for online media data; second, we present a novel approach for both discovering important event categories and classifying extracted events based on word representation and clustering model. Experimental results with real dataset show that the proposed framework is effective to extract high quality event information.

Automatic microseismic events detection using morphological multiscale top-hat transformation

The occurrence of microseismic is not random but is related to the physical properties of the underground medium.Due to the low intensity and the influence of noise,microseismic eventually lead to poor signal-to-noise ratio.We proposed a method for automatic detection of microseismic events by adoption of multiscale top-hat transformation.The method is based on the difference between the signal and noise in the multiscale top-hat transform section and achieves the detection on a specific section.The microseismic data are decomposed into different scales by multiscale morphology top-hat transformation firstly.Then the potential microseismic events could be detected by picking up the peak value in the multiscale top-hat section,and the characteristic profile obtains the start point with a specific threshold value.Finally,the synthetic data experiences demonstrate the advantages of this method under strong and weak noisy conditions,and the filed data example also shows its reliability and adaptability.

Combing Type-Aware Attention and Graph Convolutional Networks for Event Detection

Event detection(ED)is aimed at detecting event occurrences and categorizing them.This task has been previously solved via recognition and classification of event triggers(ETs),which are defined as the phrase or word most clearly expressing event occurrence.Thus,current approaches require both annotated triggers as well as event types in training data.Nevertheless,triggers are non-essential in ED,and it is time-wasting for annotators to identify the“most clearly”word from a sentence,particularly in longer sentences.To decrease manual effort,we evaluate event detectionwithout triggers.We propose a novel framework that combines Type-aware Attention and Graph Convolutional Networks(TA-GCN)for event detection.Specifically,the task is identified as a multi-label classification problem.We first encode the input sentence using a novel type-aware neural network with attention mechanisms.Then,a Graph Convolutional Networks(GCN)-based multilabel classification model is exploited for event detection.Experimental results demonstrate the effectiveness.

Automated Disabled People Fall Detection Using Cuckoo Search with Mobile Networks

Falls are the most common concern among older adults or disabled peo-ple who use scooters and wheelchairs.The early detection of disabled persons’falls is required to increase the living rate of an individual or provide support to them whenever required.In recent times,the arrival of the Internet of Things(IoT),smartphones,Artificial Intelligence(AI),wearables and so on make it easy to design fall detection mechanisms for smart homecare.The current study devel-ops an Automated Disabled People Fall Detection using Cuckoo Search Optimi-zation with Mobile Networks(ADPFD-CSOMN)model.The proposed model’s major aim is to detect and distinguish fall events from non-fall events automati-cally.To attain this,the presented ADPFD-CSOMN technique incorporates the design of the MobileNet model for the feature extraction process.Next,the CSO-based hyperparameter tuning process is executed for the MobileNet model,which shows the paper’s novelty.Finally,the Radial Basis Function(RBF)clas-sification model recognises and classifies the instances as either fall or non-fall.In order to validate the betterment of the proposed ADPFD-CSOMN model,a com-prehensive experimental analysis was conducted.The results confirmed the enhanced fall classification outcomes of the ADPFD-CSOMN model over other approaches with an accuracy of 99.17%.

Social Media Based Transportation Research: the State of the Work and the Networking

Recently, there has been an increased interest in the use of social media data as important traffic information sources.In this paper, we review social media based transportation research with social network analysis methods. We summarize main research topics in this field, and report collaboration patterns at levels of researchers, institutions, and countries, respectively.Finally, some future research directions are identified.

Distributed event region fault-tolerance based on weighted distance for wireless sensor networks

Event region detection is the important application for wireless sensor networks（WSNs）, where the existing faulty sensors would lead to drastic deterioration of network quality of service.Considering single-moment nodes fault-tolerance, a novel distributed fault-tolerant detection algorithm named distributed fault-tolerance based on weighted distance（DFWD） is proposed, which exploits the spatial correlation among sensor nodes and their redundant information.In sensor networks, neighborhood sensor nodes will be endowed with different relative weights respectively according to the distances between them and the central node.Having syncretized the weighted information of dual-neighborhood nodes appropriately, it is reasonable to decide the ultimate status of the central sensor node.Simultaneously, readings of faulty sensors would be corrected during this process.Simulation results demonstrate that the DFWD has a higher fault detection accuracy compared with other algorithms, and when the sensor fault probability is 10%, the DFWD can still correct more than 91% faulty sensor nodes, which significantly improves the performance of the whole sensor network.

Abnormal Crowd Behavior Detection Based on the Entropy of Optical Flow

To improve the detection accuracy and robustness of crowd anomaly detection,especially crowd emergency evacuation detection,the abnormal crowd behavior detection method is proposed.This method is based on the improved statistical global optical flow entropy which can better describe the degree of chaos of crowd.First,the optical flow field is extracted from the video sequences and a 2D optical flow histogram is gained.Then,the improved optical flow entropy,combining information theory with statistical physics is calculated from 2D optical flow histograms.Finally,the anomaly can be detected according to the abnormality judgment formula.The experimental results show that the detection accuracy achieved over 95%in three public video datasets,which indicates that the proposed algorithm outperforms other state-of-the-art algorithms.

Capturing semantic features to improve Chinese event detection

Current Chinese event detection methods commonly use word embedding to capture semantic representation,but these methods find it difficult to capture the dependence relationship between the trigger words and other words in the same sentence.Based on the simple evaluation,it is known that a dependency parser can effectively capture dependency relationships and improve the accuracy of event categorisation.This study proposes a novel architecture that models a hybrid representation to summarise semantic and structural information from both characters and words.This model can capture rich semantic features for the event detection task by incorporating the semantic representation generated from the dependency parser.The authors evaluate different models on kbp 2017 corpus.The experimental results show that the proposed method can significantly improve performance in Chinese event detection.

ED-SWE:Event detection based on scoring and word embedding in online social networks for the internet of people

Online social media networks are gaining attention worldwide,with an increasing number of people relying on them to connect,communicate and share their daily pertinent event-related information.Event detection is now increasingly leveraging online social networks for highlighting events happening around the world via the Internet of People.In this paper,a novel Event Detection model based on Scoring and Word Embedding(ED-SWE)is proposed for discovering key events from a large volume of data streams of tweets and for generating an event summary using keywords and top-k tweets.The proposed ED-SWE model can distill high-quality tweets,reduce the negative impact of the advent of spam,and identify latent events in the data streams automatically.Moreover,a word embedding algorithm is used to learn a real-valued vector representation for a predefined fixed-sized vocabulary from a corpus of Twitter data.In order to further improve the performance of the Expectation-Maximization(EM)iteration algorithm,a novel initialization method based on the authority values of the tweets is also proposed in this paper to detect live events efficiently and precisely.Finally,a novel automatic identification method based on the cosine measure is used to automatically evaluate whether a given topic can form a live event.Experiments conducted on a real-world dataset demonstrate that the ED-SWE model exhibits better efficiency and accuracy than several state-of-art event detection models.

An Adaptive Classifier Based Approach for Crowd Anomaly Detection

Crowd Anomaly Detection has become a challenge in intelligent video surveillance system and security.Intelligent video surveillance systems make extensive use of data mining,machine learning and deep learning methods.In this paper a novel approach is proposed to identify abnormal occurrences in crowded situations using deep learning.In this approach,Adaptive GoogleNet Neural Network Classifier with Multi-Objective Whale Optimization Algorithm are applied to predict the abnormal video frames in the crowded scenes.We use multiple instance learning(MIL)to dynamically develop a deep anomalous ranking framework.This technique predicts higher anomalous values for abnormal video frames by treating regular and irregular video bags and video sections.We use the multi-objective whale optimization algorithm to optimize the entire process and get the best results.The performance parameters such as accuracy,precision,recall,and F-score are considered to evaluate the proposed technique using the Python simulation tool.Our simulation results show that the proposed method performs better than the conventional methods on the public live video dataset.

Tracking Dengue on Twitter Using Hybrid Filtration-Polarity and Apache Flume

The world health organization(WHO)terms dengue as a serious illness that impacts almost half of the world’s population and carries no specific treatment.Early and accurate detection of spread in affected regions can save precious lives.Despite the severity of the disease,a few noticeable works can be found that involve sentiment analysis to mine accurate intuitions from the social media text streams.However,the massive data explosion in recent years has led to difficulties in terms of storing and processing large amounts of data,as reliable mechanisms to gather the data and suitable techniques to extract meaningful insights from the data are required.This research study proposes a sentiment analysis polarity approach for collecting data and extracting relevant information about dengue via Apache Hadoop.The method consists of two main parts:the first part collects data from social media using Apache Flume,while the second part focuses on querying and extracting relevant information via the hybrid filtration-polarity algorithm using Apache Hive.To overcome the noisy and unstructured nature of the data,the process of extracting information is characterized by pre and post-filtration phases.As a result,only with the integration of Flume and Hive with filtration and polarity analysis,can a reliable sentiment analysis technique be offered to collect and process large-scale data from the social network.We introduce how the Apache Hadoop ecosystem–Flume and Hive–can provide a sentiment analysis capability by storing and processing large amounts of data.An important finding of this paper is that developing efficient sentiment analysis applications for detecting diseases can be more reliable through the use of the Hadoop ecosystem components than through the use of normal machines.

A NOVEL FRAMEWORK FOR SOCCER GOAL DETECTION BASED ON SEMANTIC RULE

Focusing on the problem of goal event detection in soccer videos,a novel method based on Hidden Markov Model(HMM) and the semantic rule is proposed.Firstly,a HMM for a goal event is constructed.Then a Normalized Semantic Weighted Sum(NSWS) rule is established by defining a new feature of shots,semantic observation weight.The test video is detected based on the HMM and the NSWS rule,respectively.Finally,a fusion scheme based on logic distance is proposed and the detection results of the HMM and the NSWS rule are fused by optimal weights in the decision level,obtaining the final result.Experimental results indicate that the proposed method achieves 96.43% precision and 100% recall,which shows the effectiveness of this letter.

Data Augmentation Based Event Detection

Supervised models for event detection usually require large-scale human-annotated training data,especially neural models.A data augmentation technique is proposed to improve the performance of event detection by generating paraphrase sentences to enrich expressions of the original data.Specifically,based on an existing human-annotated event detection dataset,we first automatically build a paraphrase dataset and label it with a designed event annotation alignment algorithm.To alleviate possible wrong labels in the generated paraphrase dataset,a multi-instance learning(MIL)method is adopted for joint training on both the gold human-annotated data and the generated paraphrase dataset.Experimental results on a widely used dataset ACE2005 show the effectiveness of our approach.

Event Normalization Through Dynamic Log Format Detection

The analytical and monitoring capabilities of central event re-positories, such as log servers and intrusion detection sys-tems, are limited by the amount of structured information ex-tracted from the events they receive. Diverse networks and ap-plications log their events in many different formats, and this makes it difficult to identify the type of logs being received by the central repository. The way events are logged by IT systems is problematic for developers of host-based intrusion-detection systems （specifically, host-based systems）, develop-ers of security-information systems, and developers of event-management systems. These problems preclude the develop-ment of more accurate, intrusive security solutions that obtain results from data included in the logs being processed. We propose a new method for dynamically normalizing events into a unified super-event that is loosely based on the Common Event Expression standard developed by Mitre Corporation. We explain how our solution can normalize seemingly unrelat-ed events into a single, unified format.

A Survey on Event Tracking in Social Media Data Streams

Social networks are inevitable parts of our daily life,where an unprecedented amount of complex data corresponding to a diverse range of applications are generated.As such,it is imperative to conduct research on social events and patterns from the perspectives of conventional sociology to optimize services that originate from social networks.Event tracking in social networks finds various applications,such as network security and societal governance,which involves analyzing data generated by user groups on social networks in real time.Moreover,as deep learning techniques continue to advance and make important breakthroughs in various fields,researchers are using this technology to progressively optimize the effectiveness of Event Detection(ED)and tracking algorithms.In this regard,this paper presents an in-depth comprehensive review of the concept and methods involved in ED and tracking in social networks.We introduce mainstream event tracking methods,which involve three primary technical steps:ED,event propagation,and event evolution.Finally,we introduce benchmark datasets and evaluation metrics for ED and tracking,which allow comparative analysis on the performance of mainstream methods.Finally,we present a comprehensive analysis of the main research findings and existing limitations in this field,as well as future research prospects and challenges.

Label-Aware Chinese Event Detection with Heterogeneous Graph Attention Network

Event detection(ED)seeks to recognize event triggers and classify them into the predefined event types.Chinese ED is formulated as a character-level task owing to the uncertain word boundaries.Prior methods try to incorpo-rate word-level information into characters to enhance their semantics.However,they experience two problems.First,they fail to incorporate word-level information into each character the word encompasses,causing the insufficient word-charac-ter interaction problem.Second,they struggle to distinguish events of similar types with limited annotated instances,which is called the event confusing problem.This paper proposes a novel model named Label-Aware Heterogeneous Graph Attention Network(L-HGAT)to address these two problems.Specifically,we first build a heterogeneous graph of two node types and three edge types to maximally preserve word-character interactions,and then deploy a heterogeneous graph attention network to enhance the semantic propagation between characters and words.Furthermore,we design a pushing-away game to enlarge the predicting gap between the ground-truth event type and its confusing counterpart for each character.Experimental results show that our L-HGAT model consistently achieves superior performance over prior competitive methods.

An Efficient Method for Cleaning Dirty-Events over Uncertain Data in WSNs

Event detection in wireless sensor networks （WSNs） has attracted much attention due to its importance in many applications. The erroneous abnormal data generated during event detection are prone to lead to false detection results. Therefore, in order to improve the reliability of event detection, we propose a dirty-event cleaning method based on spatio-temporal correlations among sensor data. Unlike traditional fault-tolerant approaches, our method takes into account the inherent uncertainty of sensor measurements and focuses on the type of directional events. A probabilitybased mapping scheme is introduced, which maps uncertain sensor data into binary data. Moreover, we give formulated definitions of transient dirty-event （TDE） and permanent dirty-event （PDE）, which are cleaned by a novel fuzzy method and a collaborative cleaning scheme, respectively. Extensive experimental results show the effectiveness of our dirty-event cleaning method.

Novel Indicators for Adverse Glycemic Events Detection Analysis Based on Continuous Glucose Monitoring Neural Network Predictive Models

This paper proposes five indicators to evaluate the effectiveness and viability for adverse glycemic events detection based on predicted blood glucose(BG)values.False negative rate(FNR)and false positive rate(FPR)are defined to evaluate whether it can detect adverse glycemic events(AGEs)based on the predicted value.The temporal overlap(TO)and time difference(TD)are proposed to evaluate whether the predicted model can capture the accurate time duration of AGEs.The sum of squared percent(SSP)measures comprehensive similarity between prediction values and true values.We examined 328 patients with type 2 diabetes,containing real continuous glucose monitoring data with 5-minute time intervals.Autoregressive integrated moving average model has lower FNR and FPR.The gated recurrent unit has better temporal behavior where the mean TO with standard deviation is calculated as 0.84±0.18,and the mean TD with standard deviation is(4.39±4.01)min.Neural models have better effects on SSP(for hypoglycemia,long-short tern memory possesses 0.149 and 0.246).These five indicators are able to evaluate whether we can detect abnormal BG levels and reveal the temporal behavior of AGEs effectively.The proposed neural predictive models have more promising application in AGE detection.