Recognising situations in context aware systems using dempster shafer theory
This presentation is the property of its rightful owner.
Sponsored Links
1 / 51

Recognising Situations in context aware systems using Dempster-Shafer Theory PowerPoint PPT Presentation


  • 109 Views
  • Uploaded on
  • Presentation posted in: General

Recognising Situations in context aware systems using Dempster-Shafer Theory. Dr. Susan McKeever Nov 4 th 2013. Context Aware systems – e.g . Smart home. Sensors in a smart home Situation tracking – what is the user doing? What activity are they undertaking? E.g Monitoring elderly.

Download Presentation

Recognising Situations in context aware systems using Dempster-Shafer Theory

An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Presentation Transcript


Recognising situations in context aware systems using dempster shafer theory

Recognising Situations in context aware systemsusing Dempster-Shafer Theory

Dr. Susan McKeever

Nov 4th 2013


Recognising situations in context aware systems using dempster shafer theory

Context Aware systems – e.g. Smart home

  • Sensors in a smart home

    • Situation tracking – what is the user doing? What activity are they undertaking?

  • E.g Monitoring elderly


Recognising situations in context aware systems using dempster shafer theory

Context Aware systems

  • Pervasive /ubiquitious /ambient systems – embedded in the environment s

    • E.g. intelligent homes, location tracking system

  • They understand their own “context”.

    • Context-awarenessis the ability to track the state of the environment in order to identify situations

  • Situationsare human understandable representations of the environment, derived from sensor data


Recognising situations in context aware systems using dempster shafer theory

Research focus: e.g .Gator Tech Smart home


Recognising situations in context aware systems using dempster shafer theory

Van Kasteren sensored smart home

14 digital sensors

For a month:

7 Situations:

Preparing breakfast, dinner, drink, leave house, use toilet, take shower, go to bed


Recognising situations in context aware systems using dempster shafer theory

Abstracting sensor data to situations

Situations

Application e.g. elderly alert system

Johnis ‘preparing meal’

Is evidence of

Abstracted Context

John located in Kitchen @ time 12:30

Is abstracted to

Sensor 1, 2, 3

Sensor 1, 2, 3

Sensor 1, 2, 3

Location sensor reading

(X,Y,Z, ID239, 12:30:04)


Recognising situations in context aware systems using dempster shafer theory

Situation Recognition

Situation(s) occurring at time, t

12:53 preparing breafast

Sensor data

Situation Recognition

(12:53, 0)

(2.15,5.04,3.16, 12:34)

  • Situation recognitionis a critical, continuous, dynamic process – often required in real time.

  • The recognition process is difficult and uncertain – no single approach suitable for all

  • Knowledge

    • Expert? Past data?


Recognising situations in context aware systems using dempster shafer theory

Situation Recognition - Scenario

Scenario

“The person is in the kitchen. It is morning time. They carry out a series of tasks, such as taking cereal out of the groceries cupboard, using the kettle, opening the fridge, and using the toaster”

  • Human Observer: “preparing breakfast”

  • Why?

  • Individual tasks may not confirm that breakfast is in progress, but together, indicate the ’preparing breakfast’ situation.

  • Morning time

  • Informative sensors e.g. toaster


Recognising situations in context aware systems using dempster shafer theory

Recognising situations – Automated

Sensor overlap - Kettle and fridge: ’preparing drink?

Gaps of seconds or minutes occuring with no sensor activity – classify?

As more tasks are done, system is more certain of ‘preparing breakfast situation’ – Temporal aspect

Sensors can breakdown and have error rate – toaster sensor doesn’t fire?

The person does not prepare breakfast in the same way every day.

The tasks are not necessarily performed in any particular order.

Co-occurring situations? (’on telephone’); Cannot o-occur (’user asleep’)? -Valid combinations of situations.

Different people “prepare breakfast” in different ways.. Individual efinitions?

A second occupant now enters the kitchen – how to distinguish?


Recognising situations in context aware systems using dempster shafer theory

Recognising situations – Some approaches

  • Machine learning techniques, inc.

    • Bayesian networks

    • Decision trees

    • Hidden Markhov models

    • reliant on training data

  • Specification based approaches, inc.

    • Logic approaches

    • Fuzzy logic

    • Temporal logic


Recognising situations in context aware systems using dempster shafer theory

Problems to be solved (not exhaustive)

  • How to recognise situations in pervasive environments, allowing for particular challenges:

    • Uncertainty (sensor data, situation definitions, context fuzziness)

    • Difficulties in obtaining training data

  • My solution: Use and enhance evidence theory (Dempster Shafer theory)


Recognising situations in context aware systems using dempster shafer theory

Why Dempster Shafer theory

  • Devised in 1970s

  • Mathematical theory for combining separate pieces of information (evidence) to calculate the belief in an event.

  • Applied in military applications, cartography, image processing, expert systems, risk management, robotics and medical diagnosis

  • Key features:

    • its ability to specifically quantify and preserve uncertainty

    • its facility for assigning evidence to combinations

    • Various researchers applying in pervasive systems


Recognising situations in context aware systems using dempster shafer theory

Approach

  • Apply Dempster Shafer (evidence) theory to situation recognition

    • Create a network structure to propagate evidence from sensors

  • Extend the theory to allow for:

    • New operations needed support evidence processing of situation

    • Temporal features of situation

    • Rich (static and dynamic) sensor quality


Recognising situations in context aware systems using dempster shafer theory

Dempster Shafer theory: Example

Two sensors are used to detect user location in an office.

The locations of interest are:

(1) Cafe, (2) the user’s desk, (3) the meeting room and (4) ‘lobby’ in the building.

Meeting room

Café

User’s desk

Lobby

Frame of Discernment

‘hypotheses’

(allows combinations)

Evidence sources

Each sensors assigns belief as a ‘mass function’ which totals per sensor to 1

Sensor 1

Sensor 2

Any uncertainty is assigned to ‘ignorance’ hypthesis 𝞱–

{desk ^ cafe ^ meetingRoom ^ lobby}


Recognising situations in context aware systems using dempster shafer theory

Dempster Shafer theory: Example

Sensor 1

Detects the user’s location in the cafe.

The sensor is 70% reliable, so its belief is assigned across the frame as {cafe 0:7; 0:3 𝞱)

Sensor 2

The second sensor has conflicting evidence, assigning

{meetingRoom 0:2, desk ^cafe^lobby 0:6, 0:2 𝞱}

mass functions

To combine evidence source:

Use dempster combination rule


Recognising situations in context aware systems using dempster shafer theory

Dempster Shafer theory:

Combination rule

M12 (A) is the combination of two evidence sources or mass functions for a hypotheses A.

Denominator is a normalisation factor 1-K where K = conflicting evidence

Evidence sources must sum to 1:


Recognising situations in context aware systems using dempster shafer theory

Dempster Shafer theory: example

Sensor 1

Sensor 2

Combined evidence

Conflict (K ) = 0.14 ;

All evidence is normalised by 1-K giving:

Café 0.65; meeting 0.07; desk/café/lobby 0.21, uncertainty 0.07


Recognising situations in context aware systems using dempster shafer theory

Dempster Shafer theory: problems

Zadeh’s paradox

Conflicting sensor: Appear to agree completely if any agreement – not intuitive


Recognising situations in context aware systems using dempster shafer theory

Dempster Shafer theory: problems

Single sensor dominance

A single sensor can overrule a majority of agreeing sensors if it disagrees:

e.G .if 5 sensors determine a user location in a house, a single “categorical” (certain) sensor that assigns all its belief to a contradictory option will negate the evidence from the remaining 4.

Kitchen 0.9

Sitting room

1

Kitchen 0.6

Kitchen 0.8

Kitchen 0.7

Sensor 2

Sensor 3

Sensor 4

Sensor5

Sensor 1


Recognising situations in context aware systems using dempster shafer theory

Dempster Shafer theory: gaps

Timeline

Prepare Dinner

40 minutes

Plates

Cupboard

Accessed

No support for evidence spread over time.

Assumes evidence is all co-occuring but in reality evidencemay be spread over time.

e.g. detecting “prepare dinner” situation detected by sensors on cupboards and fridges.

Fridge

Accessed

Pans

Cupboard

Accessed

Freezer

Accessed

Groceries

Cupboard

Accessed


Recognising situations in context aware systems using dempster shafer theory

Dempster Shafer theory: gaps

Only deals with fusing evidence: no “theory” for propogating evidence across other rules in order to recognise situations

Limited to just combining n “sources”: Need a set of additional mathemtical operations for propogating evidence

Situations

Situations

Situations

Johnis ‘preparing meal’

Is evidence of

Abstracted Context

Abstracted Context

Abstracted Context

John located in Kitchen @ time 12:30

Is abstracted to

Sensor 1, 2, 3

Sensor 1, 2, 3

Sensor 1, 2, 3

Sensor 1, 2, 3

Sensor 1, 2, 3

Sensor 1, 2, 3

Sensor 1, 2, 3

Sensor 1, 2, 3

Sensor 1, 2, 3

Location sensor reading

(X,Y,Z, ID239, 12:30:04)


Recognising situations in context aware systems using dempster shafer theory

Dempster Shafer theory: gaps

Only deals with fusing evidence: no “theory” for propogating evidence across other rules in order to recognise situations (and a way to capture all this knowledge)

Situation

Situation

situation

situation

situation

Situations

Certainty

0.n

Certainty

0.n

Certainty

0.n

Context

Value

Context

Value

Context

Value

Context

Value

Context

Value

Context

Value

Abstracted Context

Context

Value

Sensor

sensor

sensor

sensor

Sensor

sensor

Sensor Level


Recognising situations in context aware systems using dempster shafer theory

Recognising situations – Using Dempster Shafer theory

  • Want an approach that reduces or eliminates reliance on training data. OK (provided we can define mass functions to say what sensor readings mean)

  • That allows for “uncertainty” OK

  • That allows temporal information to be included To be added

  • That allows sensors belief to be propogated (distributed) up into situation hierachies based on “knowledge” rules To be added

  • That addresses the issue of Zadeh’s paradox and dominant sensors To be added

  • Ultimately: Develop a full decision making architecture for real time situation recognition (overleaf) To be added

  • Needed to extend Dempster Shafer theory


Recognising situations in context aware systems using dempster shafer theory

Develop a full decision making architecture for real time situation recognition using extended DS theory

Knowlege

Valid situation

combinations

Applicati-ons

Recognised

Situations

Belief Distribution

Decision

Stage

Extended DS theory

Sensor Readings

Prep Breakfast 0.3,

Take a shower 0.6

At time t


Recognising situations in context aware systems using dempster shafer theory

Knowledge: an interconnected hierarchy of sensor and situations

Situation

Situation

situation

situation

situation

Situations

Certainty

0.n

Certainty

0.n

Certainty

0.n

Context

Value

Context

Value

Context

Value

Context

Value

Context

Value

Context

Value

Abstracted Context

Context

Value

Sensor

sensor

sensor

sensor

Sensor

sensor

Sensor Level


Recognising situations in context aware systems using dempster shafer theory

VanKasteren e.g. 3 of the situations

Get

Drink

Prepare

Breakfast

Prepare

Dinner

<2>

<15>

<62>

0.4

0.8

0.8

0.8

0.8

0.2

Morning

Nighttime

Cup

Used

Plates

Used

Microwave

Used

Groceries

Used

Freezer

Used

Pans

Used

Fridge

Used

Moning

Cup

Plates

Cupboard

Microwave

Groceries

Cupboard

Freezer

Pans

Cupboadr

Time

Fridge


Recognising situations in context aware systems using dempster shafer theory

First : Define a notation for knowledge capture : denoting sensor evidence /context/ situations –

Situation DAG

Situation

Situations

Situation

Situation

Belief distribution

< 5>

> 10 >

Certainty

0.n

Context

Values

Certainty

0.n

Certainty

0.n

Context

Value

Context

Value

Context

Value

Context

Value

Context

Value

Context

Value

Belief distribution

Sensor

sensor

Discount

0.n

Sensors


Recognising situations in context aware systems using dempster shafer theory

First : Define a notation for denoting sensor evidence /context/ situations – Situation DAG i.e to capture the knowledge of what sensors indicate what situation

is a type of

Duration of situation, evidence not in sequence

< duration>

Sensor, context value orsituation

Duration of situation, evidence in sequence

>duration >

Discount factor applied to a sensor: 0< n <1

Discount 0.n

is evidence of

Certainty 0.n

Certainty applied to an inference rule: 0 < n < 1


Recognising situations in context aware systems using dempster shafer theory

Second: Create evidence propogation rules to distribute/propogate belief up to situation level

Situation

Situation

situation

situation

situation

Situations

Up to situation certainties here

Certainty

0.n

Certainty

0.n

Certainty

0.n

Context

Value

Context

Value

Context

Value

Context

Value

Context

Value

Context

Value

Abstracted Context

Context

Value

Translate

Sensor readings into beliefs here ..

Sensor

sensor

sensor

sensor

Sensor

sensor

Sensor Level


Recognising situations in context aware systems using dempster shafer theory

Second: Create evidence propogation rules to distribute/propogate belief up to situation level

Situation

Situation

situation

situation

situation

Situations

Certainty

0.n

Certainty

0.n

Certainty

0.n

Context

Value

Context

Value

Context

Value

Context

Value

Context

Value

Context

Value

Abstracted Context

Context

Value

Sensor

sensor

sensor

sensor

Sensor

sensor

Sensor Level


Recognising situations in context aware systems using dempster shafer theory

Second: Create evidence propogation rules to distribute/propogate belief up to situation level:

Examples

Is a type of:

e.g. Situation X is occuring if either Situation Y OR Z is occuring

Occupant is “resting” if they are “watching TV” or “in bed”

Distributing combined belief across single situations


Recognising situations in context aware systems using dempster shafer theory

Second: Create evidence propogation rules to distribute/propogate belief up to situation level:

Examples: Sensor Quality

Some sensors are inherently lower quality as an evidence source

e.g. Calendar sensor is indicative of real calendar owner’s location 70% of the time – Discount (d) evidence from the sensor


Recognising situations in context aware systems using dempster shafer theory

Third: Include temporal evidence:

Timeline

Prepare Dinner

40 minutes

Fridge

Accessed

Grocery

Cupboard

accessed

Plates

Cupboard

Accessed

Freezer

Accessed

Groceries

Cupboard

Accessed

(1) Use absolute time as evidence

(2) Find a way to combine transitory evidence

Different Sensors fire intermittently – no single sensor sufficient for situation recognition


Recognising situations in context aware systems using dempster shafer theory

Third: extend evidence for duration of situation

Prepare Dinner: Time Extended Evidence

Time

Fridge

Accessed

Fridge

Extended

Fridge

Extended

Fridge

Extended

Fridge

Extended

Fridge

Extended

Groceries

Cupboard

Accessed

Groceries

Cupboard

Extended

Groceries

Cupboard

Extended

Groceries

Cupboard

Extended

Groceries

Cupboard

Extended

Plates

Cupboard

Accessed

Plates

Cupboard

Extended

Plates

Cupboard

Extended

Plates

Cupboard

Extended

Freezer

Accessed

Freezer

Extended

Freezer

Extended

Pans

Cupboard

Extended

Pans

Cupboard

Accessed

Prepare

Dinner

Starts

Prepare

Breakfast

Ends

Situation Duration


Recognising situations in context aware systems using dempster shafer theory

Fusing time extended evidence:

Adjust Dempster Shafer fusion rules to allow for time extension of evidence

Two transitory extended mass functions for hypothesis h with duration t dur, a t time t +t rem


Recognising situations in context aware systems using dempster shafer theory

Fourth: Allow for Zadeh’s and Single sensor dominance

Two options:

Use an alternative combination rule (Murphy’s) which averages out the evidence BEFORE fusing

Use a simpler averaging rule to fuse evidence

Lacks convergence

Removes Zadeh’s problem


Recognising situations in context aware systems using dempster shafer theory

Fifth: Combine all this and apply to real world data for situation recogntion

Knowlege

Valid situation

combinations

Test our approach using annotated datasets of sensor readings

Applicati-ons

Recognised

Situations

Belief Distribution

Decision

Stage

Extended DS theory

Sensor Readings

Prep Breakfast 0.3,

Take a shower 0.6

At time t


Recognising situations in context aware systems using dempster shafer theory

Experiments

  • Data set (1)

  • “Van Kasteren”

  • Heavily used by other researchers - compare results on situation recognition

  • 7 situation annotated, 14 sensors

  • Data set (2)

  • “CASL”

  • Office data set: 3 situations annotated,

  • Location sensors,

  • Calendar sensor,

  • Keyboard sensor


Recognising situations in context aware systems using dempster shafer theory

Evaluation

Various sub questions also addressed: comparison with published results, comparison of DS fusion rules, impact of quality on situation transitions, quality parameter sensitivity, static versus dynamic quality


Recognising situations in context aware systems using dempster shafer theory

Evaluation

  • 2 annotated published real world datasets – VanKasteren (Smart home) and CASL (office-based)

  • Situation DAGs created for both datasets

  • Situation recognition accuracy measured using f-measure of timesliced data sets;

  • Recognition accuracy using temporal and quality extensions evaluated

  • J45 Decision Tree and Naive Bayes used for comparison , and published results ; Cross validation used.


Recognising situations in context aware systems using dempster shafer theory

Use of DS theory with temporal extensions for situation recognition

F-Measure for each situation using DS theory – (1) no time, (2) absolute time, (3) time extended (VanKasteren dataset )


Recognising situations in context aware systems using dempster shafer theory

Temporal DS theory compared to two other approches: Naïve Bayes, J48 decision tree.

Situations


Recognising situations in context aware systems using dempster shafer theory

Our approach compared to the three available published results

Same experimental measures

*

* Excludes timeslices with no sensors firing which are harder to infer – ‘inactive’

Timeslices harder to infer


Recognising situations in context aware systems using dempster shafer theory

Use of DS theory with temporal extensions

  • Use of temporal extensions significantly improves situation accuracy (over baseline DS theory alone)

  • Performs better than J45, Naive Bayes (particularly with limited training data). This improvement narrows when more training data used (LODO)

  • Achieves 69% class accuracy in comparison to VanKasteren (49.2%) and Ye*(88.3%)


Recognising situations in context aware systems using dempster shafer theory

Use of DS theory with quality extensions

F-Measure for each situation using DS theory – with and without quality


Recognising situations in context aware systems using dempster shafer theory

Use of DS theory with quality extensions

  • Use of quality parameters significantly improves situation recognition accuracy (over baseline)

  • Performance close to Naive Bayes (4%) and J48 (2%) -

  • Each individual sensor’s quality contributes to improvement

  • Sensitivity analysis of quality parameters indicates the relative quality of sensors may be important

  • Time based dynamic quality parameters impact situation transitions – application dependant


Recognising situations in context aware systems using dempster shafer theory

Conclusions

  • Our DS theory is a viable approach to situation recognition:

  • Not reliant on training data

  • Incorporates domain knowledge

  • Caters for uncertainty

  • Encoding temporal and quality knowledge improves performance over basic DS approachBUT

  • Knowledge must be available

  • Different fusion rules appropriate in different scenarios – requires expert “evidence theory” knowledge

  • Environment changes – no feedback loop for drift

  • Potentially high computation effort can be reduced


Recognising situations in context aware systems using dempster shafer theory

Contributions

  • A situation recognition approach based on DS theory

  • Selection of existing and creation of newevidential operations and algorithms to create evidence decision networks

  • Temporal and quality extensions to DS theory

  • Diagramming technique to capture structure of evidence for an environment (Situation DAG)

  • A thorough application, evaluation and analysis of the extended DS theory approach

  • An analysis of alternative fusion rules


Recognising situations in context aware systems using dempster shafer theory

Related Publications

  • Journal

    • Journal of Pervasive and Mobile Computing

    • JAISE Volume 2, Number 2 2010

  • International Conferences

    • EuroSSC Smart Sensing UK 2009

    • ICITST Pervasive Services Italy 2008

  • International (Peer viewed)Workshops

    • Pervasive 2010, Helsinki, Finland

    • CHI 2009 Boston, US

    • QualConn 2009, Stuttgart, Germany

    • Pervasive 2009, Sydney, Australia,


Recognising situations in context aware systems using dempster shafer theory

Questions?


Recognising situations in context aware systems using dempster shafer theory

Experiments

Establish situation DAG for each dataset

-Users

-Application

experts

Situations

Context

Values

System

Developers

Sensors


  • Login