scholarly journals Analysis of Biological Record Data: Improvements of the Occupancy–Detection Models

2018 ◽  
Author(s):  
Shiyu Li
Keyword(s):  
Random Walk ◽  
List Length ◽  
Biodiversity Loss ◽  
Model Structure ◽  
Large Dataset ◽  
Prior Distributions ◽  
Running Time ◽  
Detection Model ◽  
Occupancy Detection ◽  
Record Data

AbstractMany indicators are used to monitor the progress of the target which aims to stop the biodiversity loss by 2020. However, the occupancy-detection model which is currently applied to calculate the indicator is biased. Hence, more robust models are required to track the trend of the species precisely. This research first reviews the previous works on improving this occupancy-detection model by changing the prior distributions of one of the quantities and of the models considered previously, a model based on a random walk is found to be the most appropriate although it has some potential deficiencies. Then this research provides some potential improvements of the random walk model by changing the way of modelling the prior distributions of each quantity and changing the model structure. Then the hoverflies datasets are used in this research to analyse the performance of the models. These models are compared by the running times of fitting the models and the plots of the trend of the species of all models. As a result, the categorical list length model is considered to be the most precise model among all models with a reasonable running time. Then, we fit this model with a large dataset, however, it takes a long running time to get the result. Finally, some potential improvements are suggested which may be useful for further research.

An algorithm Walktrap-SPM for detecting overlapping community structure

2017 ◽  
Vol 31 (15) ◽  
pp. 1750121 ◽  
Author(s):  
Fang Hu ◽  
Youze Zhu ◽  
Yuan Shi ◽  
Jianchao Cai ◽  
Luogeng Chen ◽  
...  
Keyword(s):  
Community Structure ◽  
Random Walk ◽  
Community Detection ◽  
Real World ◽  
Overlapping Community Detection ◽  
Overlapping Communities ◽  
Running Time ◽  
Overlapping Community ◽  
Negative Links ◽  
Novel Algorithm

In this paper, based on Walktrap algorithm with the idea of random walk, and by selecting the neighbor communities, introducing improved signed probabilistic mixture (SPM) model and considering the edges within the community as positive links and the edges between the communities as negative links, a novel algorithm Walktrap-SPM for detecting overlapping community is proposed. This algorithm not only can identify the overlapping communities, but also can greatly increase the objectivity and accuracy of the results. In order to verify the accuracy, the performance of this algorithm is tested on several representative real-world networks and a set of computer-generated networks based on LFR benchmark. The experimental results indicate that this algorithm can identify the communities accurately, and it is more suitable for overlapping community detection. Compared with Walktrap, SPM and LMF algorithms, the presented algorithm can acquire higher values of modularity and NMI. Moreover, this new algorithm has faster running time than SPM and LMF algorithms.

scholarly journals Wearable Sensor-Based Location-Specific Occupancy Detection in Smart Environments

2018 ◽  
Vol 2018 ◽  
pp. 1-21
Author(s):  
Md Abdullah Al Hafiz Khan ◽  
Nirmalya Roy ◽  
H. M. Sajjad Hossain
Keyword(s):  
Large Scale ◽  
Wearable Sensors ◽  
Real Life ◽  
Estimation Algorithm ◽  
Detection Accuracy ◽  
Detection Model ◽  
Localization Scheme ◽  
Occupancy Detection ◽  
Multimodal Data Fusion ◽  
Natural Settings

Occupancy detection helps enable various emerging smart environment applications ranging from opportunistic HVAC (heating, ventilation, and air-conditioning) control, effective meeting management, healthy social gathering, and public event planning and organization. Ubiquitous availability of smartphones and wearable sensors with the users for almost 24 hours helps revitalize a multitude of novel applications. The inbuilt microphone sensor in smartphones plays as an inevitable enabler to help detect the number of people conversing with each other in an event or gathering. A large number of other sensors such as accelerometer and gyroscope help count the number of people based on other signals such as locomotive motion. In this work, we propose multimodal data fusion and deep learning approach relying on the smartphone’s microphone and accelerometer sensors to estimate occupancy. We first demonstrate a novel speaker estimation algorithm for people counting and extend the proposed model using deep nets for handling large-scale fluid scenarios with unlabeled acoustic signals. We augment our occupancy detection model with a magnetometer-dependent fingerprinting-based localization scheme to assimilate the volume of location-specific gathering. We also propose crowdsourcing techniques to annotate the semantic location of the occupant. We evaluate our approach in different contexts: conversational, silence, and mixed scenarios in the presence of 10 people. Our experimental results on real-life data traces in natural settings show that our cross-modal approach can achieve approximately 0.53 error count distance for occupancy detection accuracy on average.

scholarly journals A New Outlier Detection Model Using Random Walk on Local Information Graph

IEEE Access ◽  
2018 ◽  
Vol 6 ◽  
pp. 75531-75544 ◽  
Author(s):  
Chao Wang ◽  
Hui Gao ◽  
Zhen Liu ◽  
Yan Fu
Keyword(s):  
Random Walk ◽  
Outlier Detection ◽  
Local Information ◽  
Detection Model ◽  
Information Graph

scholarly journals Hate Speech in the Arab Electronic Press and Social Networks

2021 ◽  
Vol 35 (6) ◽  
pp. 457-465
Author(s):  
Widad Awane ◽  
El Habib Ben Lahmar ◽  
Ayoub El Falaki
Keyword(s):  
Social Networking ◽  
Social Networking Sites ◽  
Hate Speech ◽  
Communication Tool ◽  
Large Dataset ◽  
Large Space ◽  
Detection Model ◽  
Machine Learning Methods ◽  
Open World ◽  
Combined Use

Nowadays we are witnessing an open world, characterized by globalization which is accompanied by a technology through which information circulates without borders, especially with the widespread use of social networking sites being the most common communication tool, that gives access through various applications to a large space for the presentation of multiple ideas, including extremist ideas, and the spread of hate speech. This paper introduces a system of detection of hate speech in the texts of Arabic read media and social media, which is based on a combined use of NLP, and machine learning methods. The training of the detection model is done on a large Dataset of articles, tweets and comments, collected, balanced and tokenized afterwards using BERT in Arabic. The trained model detects hate speech in Arabic and various Arabic based dialects, by classifying the texts into two classes: Neutral and Abusive. The above-mentioned model is evaluated using precision metrics, recall and f1 score, it has reached an accuracy of 83%.

scholarly journals Exact sampling of the infinite horizon maximum of a random walk over a nonlinear boundary

2019 ◽  
Vol 56 (01) ◽  
pp. 116-138
Author(s):  
Jose Blanchet ◽  
Jing Dong ◽  
Zhipeng Liu
Keyword(s):  
Random Walk ◽  
Steady State ◽  
Service Time ◽  
Time Distribution ◽  
Nonlinear Boundary ◽  
Infinite Horizon ◽  
Simulation Method ◽  
Service Time Distribution ◽  
Departure Process ◽  
Running Time

AbstractWe present the first algorithm that samples maxn≥0{Sn − nα}, where Sn is a mean zero random walk, and nα with $\alpha \in ({1 \over 2},1)$ defines a nonlinear boundary. We show that our algorithm has finite expected running time. We also apply this algorithm to construct the first exact simulation method for the steady-state departure process of a GI/GI/∞ queue where the service time distribution has infinite mean.

scholarly journals An Automatic Knee Osteoarthritis Diagnosis Method Based on Deep Learning: Data from the Osteoarthritis Initiative

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Yifan Wang ◽  
Xianan Wang ◽  
Tianning Gao ◽  
Le Du ◽  
Wei Liu
Keyword(s):  
Large Scale ◽  
Severity Assessment ◽  
Classification Result ◽  
Large Dataset ◽  
Detection Model ◽  
Semiautomatic Segmentation ◽  
Segmentation Methods ◽  
Diagnosis Method ◽  
Osteoarthritis Initiative ◽  
Learning Data

Osteoarthritis (OA) is the most common form of arthritis. According to the evidence presented on both sides of the knee bones, radiologists assess the severity of OA based on the Kellgren–Lawrence (KL) grading system. Recently, computer-aided methods are proposed to improve the efficiency of OA diagnosis. However, the human interventions required by previous semiautomatic segmentation methods limit the application on large-scale datasets. Moreover, well-known CNN architectures applied to the OA severity assessment do not explore the relations between different local regions. In this work, by integrating the object detection model, YOLO, with the visual transformer into the diagnosis procedure, we reduce human intervention and provide an end-to-end approach to automatic osteoarthritis diagnosis. Our approach correctly segments 95.57% of data at the expense of training on 200 annotated images on a large dataset that contains more than 4500 samples. Furthermore, our classification result improves the accuracy by 2.5% compared to the traditional CNN architectures.

Sign in / Sign up

Export Citation Format

Share Document