This book constitutes the thoroughly refereed post-conference proceedings of the 11th International Conference on Mobile Computing, Applications, and Services, MobiCASE 2020, held in Shanghai, China, in September 2020. The conference was held virtually due to the COVID-19 pandemic.
The 15 full papers were carefully reviewed and selected from 49 submissions. The papers are organized in topical sections on mobile application and framework; mobile application with data analysis; and AI application.
Author(s): Jing Liu, Honghao Gao, Yuyu Yin, Zhongqin Bi
Series: Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, 341
Publisher: Springer
Year: 2021
Language: English
Pages: 225
City: Cham
Preface
Organization
Contents
Mobile Application and Framework
BullyAlert- A Mobile Application for Adaptive Cyberbullying Detection
1 Introduction
2 Related Works
3 System Design and Implementation
3.1 Use Cases
3.2 Architecture and Implementation
4 User Data Analysis
5 Conclusion
References
An Optimization of Memory Usage Based on the Android Low Memory Management Mechanisms
1 Introduction
2 Related Work
3 Platform Architecture and Methodology
3.1 Improve the Memory Recovery Efficiency
3.2 Prioritize the Use of Less Memory
3.3 Prevent the Instantaneous Increase in Memory Usage
3.4 Reduce Unnecessary Memory Requests
4 Results and Discussion
4.1 Experimental Setup
4.2 Experimental Results
4.3 Discussion
5 Conclusions
References
Design of a Security Service Orchestration Framework for NFV
1 Introduction
2 NFV Architecture
2.1 Demand Analysis
3 Security Service Choreography System Architecture Design
3.1 System Workflow
3.2 AppStore Layer
3.3 Orchestration Engine Design
3.4 Virtual Security Management Platform
4 Conclusion
References
An Edge-Assisted Video Computing Framework for Industrial IoT
1 Introduction
2 Related Work
3 System Design
3.1 Discretization of Sampling Rate Distribution
3.2 Implementation of Multi-layer Channel Overlay
3.3 MultiPass Algorithm Analysis
4 Experimental Evaluation and Verification
4.1 Experimental Design
4.2 Foveal Rendering Module Evaluation
5 Conclusion
References
OBNAI: An Outlier Detection-Based Framework for Supporting Network-Attack Identification over 5G Environment
1 Introduction
2 Background
2.1 Related Works
2.2 Extreme Learning Machine
3 The Framework ODNAI
3.1 The Basic Idea
3.2 The Outlier Based Identification
3.3 The IP-Cache Algorithm
4 Experimental Evaluation
4.1 Experimental Setting
4.2 Algorithm Performance
5 Conclusions
References
Advanced Data Analysis
Image Classification of Brain Tumors Using Improved CNN Framework with Data Augmentation
1 Introduction
2 Method
2.1 Brain Tumor ROI Segmentation
2.2 Data Augmentation
2.3 Network Model Optimization
3 Experiments
3.1 Dataset and Pre-processing
3.2 Optimizer Setting
3.3 Classifier Setting
3.4 Evaluation Index
4 Analysis of Experimental Results
4.1 Effect of Parameter on Adam Optimizer
4.2 The Impact of Learning Rate and Batch Size on Model Accuracy and Loss
4.3 Accuracy Analysis of the Models Before and After Data Augmentation
4.4 Performance Analysis of Improved VGG-19 Model
4.5 Performance Analysis of Improved Inception V3 Model
4.6 Comparison with Other Methods
5 Conclusion
References
Evaluating the Effectiveness of Inhaler Use Among COPD Patients via Recording and Processing Cough and Breath Sounds from Smartphones
1 Introduction
2 Related Work
3 Data Collection
3.1 Recruitment of Subjects with COPD
3.2 Our Procedure for Recording Cough and Breath Sounds
4 Data Pre-processing
4.1 Cough and Breath Data Before Pre-processing
4.2 Pause and Noise Removal
4.3 One Second Windowing Algorithm
5 Feature Extraction & Classification Algorithms
5.1 Mel Frequency Cepstral Coefficients
5.2 Justification for MFCC Audio Features
5.3 Support Vector Machine
6 Results
6.1 Comparison of Results Using Other Algorithms
6.2 Complexity of Execution
7 Conclusions and Future Work
References
Safe Navigation by Vibrations on a Context-Aware and Location-Based Smartphone and Bracelet Using IoT
1 Introduction
2 Related Work
2.1 Information Consumption Process
2.2 Information Production Process
3 Solution Approach
4 The Vibration Language
4.1 Participants and Experiments
4.2 The Language
5 System Overview
5.1 In-Advanced Planning Routes on Mobile Phones
5.2 The Bracelet for Real-Time Navigation
5.3 Synchronization Between Mobile, Cloud and Bracelet
6 Limitations
7 Discussion and Conclusion
References
Key Location Discovery of Underground Personnel Trajectory Based on Edge Computing
1 Introduction
2 Materials and Methods
2.1 Personnel Semantics
2.2 Detection of Inflection Point
2.3 Detection of a Stay Point
3 Results and Analysis
4 Conclusions
References
An Improved Spectral Clustering Algorithm Using Fast Dynamic Time Warping for Power Load Curve Analysis
1 Introduction
2 Model Building
2.1 Similarity Measure
2.2 Clustering Algorithms
3 Experiment and Analysis
3.1 Basic Description of the Experimental Environment
3.2 Clustering Quality Evaluation Index
3.3 Experiment Analysis
4 Conclusion
References
AI Application
Research on Text Sentiment Analysis Based on Attention CMGU
1 Introduction
2 Recurrent Neural Network with Gate Structure
2.1 LSTM and GRU
2.2 Minimum Gated Unit MGU
3 Sentiment Analysis Model Based on Attention CMGU
3.1 Model Design
3.2 Algorithm Design
4 Experimental Analysis
4.1 Data Set
4.2 Experimental Settings
4.3 Analysis of Experimental Results
5 Conclusions
References
Inception Model of Convolutional Auto-encoder for Image Denoising
1 Introduction
2 Network Structure of Convolutional Auto-encoder
2.1 Generate Noise Image
2.2 Multi Feature Extraction Inception Module
2.3 Design of Convolutional Auto-Encoder Network Based on Inception Module
3 Experiments and Results
3.1 Experimental Data
3.2 Experimental Environment and Evaluation Criteria
3.3 Influence of Network Model on Denoising Performance
3.4 Comparative Experimental Analysis
4 Epilogue
References
Detection and Segmentation of Graphical Elements on GUIs for Mobile Apps Based on Deep Learning
1 Introduction
2 Related Work
3 Methodology
3.1 Dataset Creation
3.2 Mask R-CNN for Detection and Segmentation
4 Experimental Results
4.1 Training Environment
4.2 Evaluation
5 Conclusion
References
Approximate Sub-graph Matching over Knowledge Graph
1 Introduction
2 Background
2.1 Related Work
2.2 Problem Definition
3 Skyline Based Sub-RDF Matching
3.1 The Compression Algorithm
3.2 The Sub-graph Matching Algorithm
3.3 The Skyline Algorithm
4 Experimental Study
4.1 Experiment Setup
4.2 Algorithm Performance
5 Conclusions
References
Metamorphic Testing for Plant Identification Mobile Applications Based on Test Contexts
1 Introduction
2 Related Work
2.1 Metamorphic Testing
2.2 Testing and Verification of AI-Based Software Systems
3 Approach
3.1 Test Context Construction
3.2 Metamorphic Testing for Plant Identification Apps
3.3 Follow-Up Image Generation
4 Case Study
4.1 Dataset
4.2 Plant Identification Apps Under Test
4.3 Evaluation Metrics
4.4 Case Study Results
4.5 Threats to Validity
5 Conclusion and Future Work
References
Author Index
