Deep Learning, Ubiquitous and Toy Computing Minitrack

Permanent URI for this collection

The pervasive nature of digital technologies as witnessed in industry, services and everyday life has given rise to an emergent, data-focused economy stemming from many aspects of human individual and ubiquitous applications. The richness and vastness of these data are creating unprecedented research opportunities in a number of fields including urban studies, geography, economics, finance, entertainment, and social science, as well as physics, biology and genetics, public health and many other smart devices. In addition to data, text and machine mining research, businesses and policy makers have seized on deep learning technologies to support their decisions and proper growing smart application needs.

As businesses build out emerging hardware and software infrastructure, it becomes increasingly important to anticipate technical and practical challenges and to identify best practices learned through experience in this research area. Deep learning employs software tools from advanced analytics disciplines such as data mining, predictive analytics, text and machine learning based on a set of algorithms that attempt to model high-level abstractions in data by using multiple processing layers with complex structures or non-linear transformations.

At the same time, the processing and analysis of deep learning applications present methodological and technological challenges. Further deep learning applications are advantaged by a rise in sensing technologies as witnessed in both the number of sensors and the rich diversity of sensors ranging from cell phones, personal computers, and health tracking appliances to Internet of Things (IoT) technologies designed to give contextual, semantic data to entities in an ubiquitous environment that previously could not contribute intelligence to key decisions and smart devices. Recently deep learning technologies have been applied into toy computing. Toy computing is a recently developing concept which transcends the traditional toy into a new area of computer research using ubiquitous technologies. A toy in this context can be effectively considered a computing device or peripheral called Smart Toys. We invite research and industry papers related to these specific challenges and others that are driving innovation in deep learning, ubiquitous and toy computing.

The goal of this minitrack is to present both novel and industrial solutions to challenging technical issues as well as compelling smart application use cases. This minitrack will share related practical experiences to benefit the reader, and will provide clear proof that deep learning technologies are playing an ever-increasing important and critical role in supporting ubiquitous and toy computing applications - a new cross-discipline research topic in computer science, decision science, and information systems. With a general focus on deep learning, ubiquitous and toy computing, this minitrack covers related topics in deep learning, ubiquitous and toy computing such as:

  • Data Modeling and Implementation
  • Analytics and Algorithms
  • Business Models
  • Delivery, Deployment and Maintenance
  • Real-time Processing Technologies and Online Transactions
  • Conceptual and Technical Architecture
  • Visualization Technologies
  • Modeling and Implementation
  • Security, Privacy and Trust
  • Industry Standards and Solution Stacks
  • Provenance Tracking Frameworks and Tools
  • Software Repositories
  • Organizations Best Practices
  • Case Studies (e.g., smart toys, healthcare, financial, aviation, etc.)

Minitrack Chair:

Patrick C. K. Hung (Primary Contact)
Faculty of Business and Information Technology, University of Ontario Institute of Technology, Canada
Department of Electronic Engineering, National Taipei University of Technology, Taiwan
Email: patrick.hung@uoit.ca

Shih-Chia Huang
Department of Electronic Engineering, National Taipei University of Technology, Taiwan
Email: schuang@ntut.edu.tw

Sarajane Marques Peres
School of Arts, Sciences and Humanities, University of São Paulo, Brazil
Email: sarajane@usp.br​

Browse

Recent Submissions

Now showing 1 - 4 of 4
  • Item
    Towards a Privacy Rule Conceptual Model for Smart Toys
    ( 2017-01-04) Rafferty, Laura ; Hung, Patrick ; Fantinato, Marcelo ; Marques Peres, Sarajane ; Iqbal, Farkhund ; Kuo, Sy-Yen ; Huang, Shih-Chia
    A smart toy is defined as a device consisting of a physical toy component that connects to one or more toy computing services to facilitate gameplay in the cloud through networking and sensory technologies to enhance the functionality of a traditional toy. A smart toy in this context can be effectively considered an Internet of Things (IoT) with Artificial Intelligence (AI) which can provide Augmented Reality (AR) experiences to users. In this paper, the first assumption is that children do not understand the concept of privacy and the children do not know how to protect themselves online, especially in a social media and cloud environment. The second assumption is that children may disclose private information to smart toys and not be aware of the possible consequences and liabilities. This paper presents a privacy rule conceptual model with the concepts of smart toy, mobile service, device, location, and guidance with related privacy entities: purpose, recipient, obligation, and retention for smart toys. Further the paper also discusses an implementation of the prototype interface with sample scenarios for future research works.
  • Item
    Excuse Me, Do I Know You From Somewhere? Unaware Facial Recognition Using Brain-Computer Interfaces
    ( 2017-01-04) Bellman, Christopher ; Vargas Martin, Miguel ; Liscano, Ramiro ; Alomari, Ruba ; MacDonald, Shane
    While a great deal of research has been done on \ the human brain’s reaction to seeing faces and \ reaction to recognition of these faces, the unaware \ recognition of faces is an area where further research \ can be conducted and contributed to. We performed a \ preliminary experiment where participants viewed \ images of faces of individuals while we recorded their \ EEG signals using a consumer-grade BCI headset. \ Pre-selection of the images used in each of the three \ phases in the experiment allowed us to tag each image \ based on what state of recognition we expect the image \ to take – No Recognition, a Possible Unaware \ Recognition, and a Possible Aware Recognition. We \ find, after filtering, artifact removal, and analysis of \ the participants’ EEG signals recorded from a \ consumer-grade BCI headset, obvious differences \ between the three classes of recognition (as defined \ above) and, more specifically, unaware recognitions, \ can be easily identified.
  • Item
    Effective Matrix Factorization for Online Rating Prediction
    ( 2017-01-04) Zhou, Bowen ; Wong, Raymond
    Recommender systems have been widely utilized by online merchants and online advertisers to promote their products in order to improve profits. By evaluating customer interests based on their purchase history and relating it to commodities for sale these retailers could excavate out products which are most likely to be chosen by a specific customer. In this case, online ratings given by customers are of great interest as they could reflect different levels of customers’ interest on different products. Collaborative Filtering (CF) approach is chosen by a large amount of web-based retailers for their recommender systems because CF operates on interactions between customers and products. In this paper, a major approach of CF, Matrix Factorization, is modified to give more accurate recommendations by predicting online ratings.
  • Item
    Introduction to Deep Learning, Ubiquitous and Toy Computing Minitrack
    ( 2017-01-04) Hung, Patrick ; Hunag, Shih-Chia