Why we do not want to build intelligent mobile applications



Similar documents
GIMBAL PLATFORM DIGITAL INSIGHTS INTO THE PHYSICAL WORLD

Introducing BEEKS Proximity Solutions. Developer Kit Gets You Started

Mobile crowdsensing of parking space using geofencing and. activity recognition

CASE STUDY. Enhancing the Patient Experience Harris Mobile Patient Engagement Platform

Horizontal IoT Application Development using Semantic Web Technologies

Industry 4.0 and Big Data

AN INFORMATION AGENT SYSTEM FOR CLOUD COMPUTING BASED LOCATION TRACKING

Information Systems and Technologies in Organizations

Figure 1 Cloud Computing. 1.What is Cloud: Clouds are of specific commercial interest not just on the acquiring tendency to outsource IT

Service Oriented Architecture

Clonecloud: Elastic execution between mobile device and cloud [1]

Masters in Information Technology

An Introduction to Data Mining

Application Development for Mobile and Ubiquitous Computing

Mobile and Cloud computing and SE

FITMAN Future Internet Enablers for the Sensing Enterprise: A FIWARE Approach & Industrial Trialing

Literature Review: Starting Mobile Application Development for E-Sports Portal Hayoung Noh

Proximity Marketing Privacy Considerations

Professional Organization Checklist for the Computer Science Curriculum Updates. Association of Computing Machinery Computing Curricula 2008

CHAPTER 8 CLOUD COMPUTING

Programming the Internet of Things

E-Business Technologies for the Future

Masters in Human Computer Interaction

1. The information we collect and how we collect it.

Masters in Advanced Computer Science

Masters in Artificial Intelligence

Masters in Networks and Distributed Systems

A Secure Autonomous Document Architecture for Enterprise Digital Right Management

Artificial Intelligence and Politecnico di Milano. Presented by Matteo Matteucci

Self-Protecting Documents for Cloud Storage Security

ONEM2M SERVICE LAYER PLATFORM

Cloud Computing Services and its Application

999GPS.net Tracking Platform Operation Guide

Imam Mohammad Ibn Saud Islamic University College of Computer and Information Sciences Department of Computer Sciences

VALLIAMMAI ENGNIEERING COLLEGE SRM Nagar, Kattankulathur DEPARTMENT OF COMPUTER SCIENCE AND ENGINEERING

CNG IN A BOX: Cloud Based Enterprise Historian w\dash Boarding Solution for CNG Fueling Stations

Technical Club: New Vision of Computing

M.Tech. Software Systems

ECE 455/555 Embedded System Design. Android Programming. Wei Gao. Fall

CPS221 Lecture: Cloud Computing last revised 10/22/14 Objectives

Beacon TRENDS in the Retail Space 2015

Masters in Computing and Information Technology

PROGRAM DIRECTOR: Arthur O Connor Contact: URL : THE PROGRAM Careers in Data Analytics Admissions Criteria CURRICULUM Program Requirements

School of Computer Science

HTML5. Turn this page to see Quick Guide of CTTC

Computer Science Electives and Clusters

Introduction to Engineering Using Robotics Experiments Lecture 18 Cloud Computing

Getting Started with ibeacon

About Blue Sky Sessions

A mobile monitoring and alert SMS system with remote configuration A case study for android and the fused location provider

Introduction. A. Bellaachia Page: 1

Mobile Phone & Website Tracking Platform Operation Guide

Sunnie Chung. Cleveland State University

Bachelor of Games and Virtual Worlds (Programming) Subject and Course Summaries

OPC UA App development for Android

How To Build A Cloud Computing And Healthcare Based Platform For A Mobile Health Care System

Development. SriSeshaa Technologies. Table of Contents

Present and Act Upon. Register. Consume. Stream Analytics. Event Hubs. Field Gateway. Applications Cloud Gateway. Legacy IoT (custom protocols)

Cloud Computing and Advanced Relationship Analytics

CommonTime Making Business Mobile. Enterprise. CommonTime. Mobile Solutions. mdesign Platform.

Introduction to Mobile GIS. An overview of mobile GIS technologies for fun and profit.

Doctor of Philosophy in Computer Science

Mobile App Testing Process INFLECTICA TECHNOLOGIES (P) LTD

SOLUTION BRIEF. Increase Business Agility with the Right Information, When and Where It s Needed. SAP BusinessObjects Business Intelligence Platform

The Impact of Computer Engineering 1. The Impact of Computer Engineering Oakland University Andrew Nassif 11/21/2015

On the features and challenges of security and privacy in distributed internet of things. C. Anurag Varma CpE /24/2016

School of Computer Science

Lecture Embedded System Security A. R. Darmstadt, Introduction Mobile Security

Monitis Project Proposals for AUA. September 2014, Yerevan, Armenia

The Ethics of Cloud Computing A Conceptual Review

Towards an On board Personal Data Mining Framework For P4 Medicine

What is Artificial Intelligence?

Cloud Manufacturing Olena Skarlat

Sensor Fusion Mobile Platform Challenges and Future Directions Jim Steele VP of Engineering, Sensor Platforms, Inc.

Self-Service Business Intelligence

Specialized Android APP Development Program with Java (SAADPJ) Duration 2 months

SOA, case Google. Faculty of technology management Information Technology Service Oriented Communications CT30A8901.

Cloud computing based big data ecosystem and requirements

Application of Android Mobile Platform in Remote Medical Monitoring System

Putchong Uthayopas, Kasetsart University

Transcription:

Why we do not want to build intelligent mobile applications Szymon Bobek Institute of Applied Computer Science AGH University of Science and Technology http://geist.agh.edu.pl Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 1 / 37

Outline I 1 Introduction 2 General issues 3 Engineering issues 4 Social and philosophical issues 5 Summary Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 2 / 37

Presentation Outline 1 Introduction What is (Artificial) Intelligence What are CAS How to build CAS 2 General issues 3 Engineering issues 4 Social and philosophical issues 5 Summary Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 3 / 37

What is intelligence Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 4 / 37

Can submarine swim? Artificial Intelligence A question Can machine think? is similar to a question Can submarine swim?. Artificial intelligence is just a simulation of real intelligence, hence it requires appropriate model. Every model requires constraints. The looser constraints, the more difficult the simulation. Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 5 / 37

Outline 1 Introduction What is (Artificial) Intelligence What are CAS How to build CAS 2 General issues 3 Engineering issues Gathering context Modelling context Processing context 4 Social and philosophical issues Semantics vs. context Usability, intelligibility Privacy 5 Summary Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 6 / 37

In Theory Context Where you are, who you are with, what resources are nearby (Schillit) Any informaiton that can be used to characterize the situation of an entity (Dey) Individuality, activity, location, time, relations (Zimmerman) Set of variables that may be of interest for an agent and that influence its actions (Bolchini) Aware Artificial intelligence methods Systems Intelligent homes, intelligent cars, robotics Ambient intelligence, pervasive environments, ubiquitous computing Mobile computing (location aware mobile applicaitons) Intelligent software (contextual advertising, etc.) Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 7 / 37

In Practice Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 8 / 37

In Practice Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 8 / 37

In Practice Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 8 / 37

In Practice Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 8 / 37

In Practice Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 8 / 37

In Practice Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 8 / 37

Context is not only a location Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 9 / 37

Context is not only a location Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 9 / 37

Outline 1 Introduction What is (Artificial) Intelligence What are CAS How to build CAS 2 General issues 3 Engineering issues Gathering context Modelling context Processing context 4 Social and philosophical issues Semantics vs. context Usability, intelligibility Privacy 5 Summary Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 10 / 37

Acquire, represent, use Collect Interpret, Represent, Model Process, Use Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 11 / 37

Presentation Outline 1 Introduction 2 General issues 3 Engineering issues 4 Social and philosophical issues 5 Summary Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 12 / 37

Why we do not want to build intelligent mobile applications? Because However Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 13 / 37

Why we do not want to build intelligent mobile applications? Because However Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 13 / 37

Why we do not want to build intelligent mobile applications? Because However Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 13 / 37

Why we do not want to build intelligent mobile applications? Because However Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 13 / 37

Why we do not want to build intelligent mobile applications? Because However No tools Well No skills Not possible No point EIS @ AGH Jakdojade.pl, Google now, Google car We create the point Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 13 / 37

Presentation Outline 1 Introduction 2 General issues 3 Engineering issues Gathering context Modelling context Processing context 4 Social and philosophical issues 5 Summary Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 14 / 37

Android API and SensorManager Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 15 / 37

AWARE All in one solution 24 context providers implemented Open source http://www.awareframework.com client and server solution Plug-ins philosophy (so far about 15 plug-ins) Service oriented architecture Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 16 / 37

AWARE All in one solution 24 context providers implemented Open source http://www.awareframework.com client and server solution Plug-ins philosophy (so far about 15 plug-ins) Service oriented architecture Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 16 / 37

Estimote and Gimbal Microlocation with beacons Based on Bluetooth Low Energy (BLE technology) Opposite to GPS it allows detecting device position within a building or a room Android and ios API Preorder for 99$ Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 17 / 37

Estimote and Gimbal Microlocation with beacons Based on Bluetooth Low Energy (BLE technology) Opposite to GPS it allows detecting device position within a building or a room Android and ios API Preorder for 99$ Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 17 / 37

It looks awesome, but... Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 18 / 37

It looks awesome, but... Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 18 / 37

Outline 1 Introduction What is (Artificial) Intelligence What are CAS How to build CAS 2 General issues 3 Engineering issues Gathering context Modelling context Processing context 4 Social and philosophical issues Semantics vs. context Usability, intelligibility Privacy 5 Summary Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 19 / 37

Why bother with models For the same reason we......put data into database,...use UML,...design and plan things. So we are able to......add structure to data...add semantics to meaningless data...enhance/allow/prepare data for processing...allow data exchange and system interoperability Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 20 / 37

Available modelling approaches Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 21 / 37

It looks awesome, but... Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 22 / 37

It looks awesome, but... Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 22 / 37

Outline 1 Introduction What is (Artificial) Intelligence What are CAS How to build CAS 2 General issues 3 Engineering issues Gathering context Modelling context Processing context 4 Social and philosophical issues Semantics vs. context Usability, intelligibility Privacy 5 Summary Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 23 / 37

Android API Geolocation Entering the geofence Entering and dwelling for some period of time Exiting the geofence ActivityRecognition The device is in a vehicle The device is on a bicycle The device is on a user who is walking or running. The device is still. The device angle relative to gravity changed significantly. Unable to detect the current activity. Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 24 / 37

Machine Learning (Not) a rocket science BigData ;) Weka, Matlab, Python for rapid prototyping JavaML for development Examples Clustering - for discovering patterns, groups Probabilistic graphical models - for handling uncertainty, predicting Regression - for discovering trends, patterns Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 25 / 37

AWARE Framework Framework support Sensing Processing (offline via Context Providers, or on the server side) Sharing and communicating (via MQTT messages) Binding with other applications (via Context Observers and Context Broadcasters) Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 26 / 37

It looks awesome, but... Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 27 / 37

It looks awesome, but... Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 27 / 37

Presentation Outline 1 Introduction 2 General issues 3 Engineering issues 4 Social and philosophical issues Semantics vs. context Usability, intelligibility Privacy 5 Summary Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 28 / 37

Context vs. semantics Fall detection The context is that a person is laying on the floor The semantic explains what does it mean Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 29 / 37

Context vs. semantics Fall detection The context is that a person is laying on the floor The semantic explains what does it mean Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 29 / 37

Outline 1 Introduction What is (Artificial) Intelligence What are CAS How to build CAS 2 General issues 3 Engineering issues Gathering context Modelling context Processing context 4 Social and philosophical issues Semantics vs. context Usability, intelligibility Privacy 5 Summary Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 30 / 37

Should user understand how system works? Intelligibility Ability of the system to explain how it works. Capability of being understood. The Clippy Microsoft Agent has mostly been abandoned because it made erroneous suggestions with no explanation of why these suggestions were being made Amazon.com added a link under a user s recommendations: Why is this recommended for you? Intelligibility improves usability, however only when a system is certain its decisions Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 31 / 37

Outline 1 Introduction What is (Artificial) Intelligence What are CAS How to build CAS 2 General issues 3 Engineering issues Gathering context Modelling context Processing context 4 Social and philosophical issues Semantics vs. context Usability, intelligibility Privacy 5 Summary Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 32 / 37

Mind the... user Locally or in a cloud People do not feel comfortable sharing their location, and other personal data. Cloud sounds good only to developers users prefer Dropbox, GoogleDrive. Processing large amounts of information locally costs energy Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 33 / 37

Presentation Outline 1 Introduction 2 General issues 3 Engineering issues 4 Social and philosophical issues 5 Summary Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 34 / 37

In one word Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 35 / 37

In one word Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 35 / 37

In one word Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 35 / 37

In one word Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 35 / 37

Challenges Under research Energy consumption Intelligibility and usability Processing context, adaptability Methodologies for building CAS, modelling techniques and procedures Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 36 / 37

Thank you! Szymon Bobek Institute of Applied Computer Science AGH University of Science and Technology 17 January 2014 http://geist.agh.edu.pl http://wownow.pl Szymon Bobek (AGH-UST) Mobile Trends 2014 17 January 2014 37 / 37