A Deep Learning Spatiotemporal Prediction Framework for Mobile Crowdsourced Services

@article{Said2019ADL,
  title={A Deep Learning Spatiotemporal Prediction Framework for Mobile Crowdsourced Services},
  author={Ahmed Ben Said and Abdelkarim Erradi and Azadeh Ghari Neiat and Athman Bouguettaya},
  journal={Mobile Networks and Applications},
  year={2019},
  volume={24},
  pages={1120-1133}
}
This papers presents a deep learning-based framework to predict crowdsourced service availability spatially and temporally. A novel two-stage prediction model is introduced based on historical spatio-temporal traces of mobile crowdsourced services. The prediction model first clusters mobile crowdsourced services into regions. The availability prediction of a mobile crowdsourced service at a certain location and time is then formulated as a classification problem. To determine the availability… 

Deep-Gap: A Deep Learning Framework for Forecasting Crowdsourcing Supply-Demand Gap Based on Imaging Time Series and Residual Learning

  • Ahmed Ben SaidA. Erradi
  • Computer Science
    2019 IEEE International Conference on Cloud Computing Technology and Science (CloudCom)
  • 2019
TLDR
Deep-Gap is proposed, a deep learning approach based on residual learning to predict the gap between mobile crowdsourced service supply and demand at a given time and space that achieves the lowest forecasting errors in both scenarios.

Multiview topological data analysis for crowdsourced service supply-demand gap prediction

  • Ahmed Ben SaidA. Erradi
  • Computer Science
    2020 International Wireless Communications and Mobile Computing (IWCMC)
  • 2020
TLDR
This work proposes a forecasting approach to infer the supply demand gap of crowdsourced services in a given geographic area using state-of-art prediction approach that generates predictors from the raw gap data using topological data analysis to exploit the topological and underlying geometric structures.

Deep Learning for Mobile Crowdsourcing Techniques, Methods, and Challenges: A Survey

TLDR
The current research status of deep learning for mobile crowdsourcing is reviewed from the perspectives of techniques, methods, and challenges.

A Probabilistic Approach for Maximizing Travel Journey WiFi Coverage Using Mobile Crowdsourced Services

TLDR
This work proposes a probabilistic approach to discover groups of available crowdsourced WiFi services along with the journey segments and proposes an algorithm to determine the best public transport journey plan offering based on the QoS of available WiFi service groups along the journey path.

DeepAuto: A Hierarchical Deep Learning Framework for Real-Time Prediction in Cellular Networks

TLDR
This work introduces a reusable analytics framework that enables real-time KPI prediction using a hierarchical deep learning architecture, namely DeepAuto, that stacks multiple long short-term memory networks horizontally to capture instantaneous, periodic and seasonal patterns in KPI time-series.

Machine Learning in Mobile Crowd Sourcing: A Behavior-Based Recruitment Model

TLDR
Simulations based on a real-life dataset show that considering human behavior in varying situations improves the quality of recruitment achieved by the tasks and their completion confidence when compared with a benchmark that relies solely on reputation.

FacebookVideoLive18: A Live Video Streaming Dataset for Streams Metadata and Online Viewers Locations

TLDR
A dataset gathered from one of the popular live streaming platforms: Facebook, storing more than 1,500,000 live stream records collected in June and July 2018, to collect fine grained metadata of public videos that might be useful for the research community.

On The Effectiveness of Imaging of Time Series for Flare Forecasting Problem

TLDR
This study converts multivariate time series of pre-flare magnetic fields’ activity into multi-channel image data using Gramian Angular Fields (GAF) and Markov Transition Fields (MTF) to explore the proven strength of deep neural networks in the Image Processing domain, on flares’ MVTS data.

Emerging Solutions in Big Data and Cloud Technologies for Mobile Networks

TLDR
This special issue aims to bring together researchers interested in formulating and implementing innovative solutions under big data and cloud technologies for mobile networks to the interest of interdisciplinary target audience spanning: any recent challenge applicable to mobile networks.

References

SHOWING 1-10 OF 44 REFERENCES

Spatio-Temporal Composition of Crowdsourced Services

TLDR
A new QoS-aware spatio-temporal union composition algorithm is presented to efficiently select the optimal crowdsourced composition plan for crowdsourced services based on dynamic features such as spatio -temporal aspects.

Crowdsourced Coverage as a Service: Two-Level Composition of Sensor Cloud Services

TLDR
A novel quality model is introduced that considers dynamic features of sensors to select and compose Sensor-Cloud services and defines Coverage as a Service which is formulated as a composition of crowdsourced Sensor- cloud services.

Modeling User Activity Preference by Leveraging User Spatial Temporal Characteristics in LBSNs

TLDR
A STAP model is proposed that first models the spatial and temporal activity preference separately, and then uses a principle way to combine them for preference inference, and a context-aware fusion framework is put forward to combine the temporal and spatial activity preference models for preferences inference.

Spatio-temporal Composition of Sensor Cloud Services

TLDR
This work proposes a heuristic algorithm based on A* to compose Sensor-Cloud services in terms of spatio-temporal aspects, and presents a new spatio -temporal technique based on 3D R-tree to access Sensor- cloud services.

Scalable and Accurate Prediction of Availability of Atomic Web Services

TLDR
LUCS is presented, a formal model for predicting the availability of atomic web services that enhances the current state-of-the-art models used in service recommendation systems and significantly improves availability prediction when all of the LUCS input parameters are available.

Encoding Time Series as Images for Visual Inspection and Classification Using Tiled Convolutional Neural Networks

TLDR
This work used Tiled Convolutional Neural Networks (tiled CNNs) on 12 standard datasets to learn high-level features from individual GAF, MTF, and GAF-MTF images that resulted from combining GAF and MTF representations into a single image.

Hierarchical recurrent neural network for skeleton based action recognition

  • Yong DuWei WangLiang Wang
  • Computer Science
    2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2015
TLDR
This paper proposes an end-to-end hierarchical RNN for skeleton based action recognition, and demonstrates that the model achieves the state-of-the-art performance with high computational efficiency.

Deep Learning

TLDR
Deep learning is making major advances in solving problems that have resisted the best attempts of the artificial intelligence community for many years, and will have many more successes in the near future because it requires very little engineering by hand and can easily take advantage of increases in the amount of available computation and data.