Presentation of the paper "Modelling and Managing Ambiguous Context in Intelligent Environments "in the UCAMI 2011 conference.

1. Modelling and Managing Ambiguous
Context in Intelligent Environments
Aitor Almeida, Diego López-de-Ipiña
Deusto Institute of Technology - DeustoTech
University of Deusto
Bilbao, Spain
{aitor.almeida, dipina}@deusto.es

2. Need for ambiguity modeling
• We can’t take certainty for granted when modeling real world
environments
– Unreliable sensors: Sensors usually have a known degree of precision
– Conflicting measures: What happens if various sensors of a room
provide different measures?
– Users’ subjective view of reality: the does not carry the same exact
meaning for all the users.

3. Components of the ambiguity
• We have considered two aspects of the ambiguity:
uncertainty and vagueness
• Uncertainty models the truthfulness of the different context
data by assigning to them a certainty factor
– E.g. The temperature of this room is 23 ºC with a certainty of 0.9
• Vagueness helps us to model those situations where the
boundaries between categories are not clearly defined
– E.g. The room can be cold, medium or hot.

4. AMBI2ONT: An ontology for the
ambiguous reality
• We have created an ontology that takes into account the
certainty (represented by a certainty factor, CF) and the
vagueness (represented by fuzzy sets) of the context.
– “the temperature of the room is 27ºC with a certainty factor of 0.2 and
18ºC with a certainty factor of 0.8”
– “the temperature of the room is cold with a membership of 0.7”

5. AMBI2ONT: An ontology for the
ambiguous reality

6. AMBI2ONT: An ontology for the
ambiguous reality
• Each ContextData individual has the following properties:
– crisp_value: the measure taken by the associated sensor. In our
system a sensor is defined as anything that provides context
information.
– certainty_factor: the degree of credibility of the measure. This metric
is given by the sensor that takes the measure and takes values
between 0 and 1.
– linguistic_term: each measure has its fuzzy
representation, represented as the linguistic term name and the
membership degree for that term.

7. AMBI2ONT: An ontology for the
ambiguous reality

8. Semantic Context Management For
Ambiguous Data

9. Semantic Context Management For
Ambiguous Data
1. Adding the measure
– The sensor must provide the measure type, its value, location and a
certainty factor
2. Processing the semantic and spatial data:
– We apply a semantic inference process to achieve two goals:
• make explicit the hidden implicit knowledge in the ontology
• infer the positional information of each measure
– Two different sets of rules: the semantic rules and the spatial
heuristic rules

10. Semantic Context Management For
Ambiguous Data
• Example of semantic rules:

11. Semantic Context Management For
Ambiguous Data
• Example of spatial rules:

12. Semantic Context Management For
Ambiguous Data
3. The data fusion process
– Each room can have multiple sensors that provide the same context data
(e.g various thermometers in the same room).
– The values and certainty factor of those measures can differ.
– The process refines those individual measures into a single global
measure for each room.
– We use two types of strategies for this process: tourney and
combination.
– Using the tourney strategy the measure with the best CF is selected as
the global measure of the room.
– The combination strategy has three different behaviors :
• Severe: The worst certainty factor from all the input measures is assigned to the
combined measure.
• Indulgent: The best certainty factor from all the input measures is assigned to the
combined measure.
• Cautious: An average certainty factor is calculated using the certainty factor from the
input measures.

13. Semantic Context Management For
Ambiguous Data
3. The data fusion process
– To determine the combined measure value we weight the individual
values:
Where:
m: the measure values.
cf: the measure certainty factor.

14. Semantic Context Management For
Ambiguous Data
4. Processing the ambiguity
– We have modified the JFuzzyLogic Open Source fuzzy reasoner to
accept also uncertainty information
– The modified reasoner supports two types of uncertainty, uncertain
data and uncertain rules.
– The first type occurs when the input data is not completely reliable

15. Semantic Context Management For
Ambiguous Data
4. Processing the ambiguity
– The second type of uncertainty takes place when the outcome of a
rule is not fixed
• E.g. “if the barometric pressure is high and the temperature is low there is a 60%
chance of rain”

16. Semantic Context Management For
Ambiguous Data
4. Processing the ambiguity
– Uncertainty and fuzziness can appear in the same rule and influence
each other.
• To process this we use the model described in “R. A. Orchard. FuzzyCLIPS Version
6.04A User’s Guide. Integrated Reasoning Institute for Information Technology
National Research Council Canada. 1998”.
– This model contemplates three different situations:
• CRISP Simple Rule where both antecedent and matching fact are crisp values,
• FUZZY_CRISP Simple Rule where both the antecedent and matching fact are fuzzy
and the consequent is crisp and finally the
• FUZZY_FUZZY Simple rule where all three are fuzzy.

17. Semantic Context Management For
Ambiguous Data
4. Processing the ambiguity
– CF in CRISP Simple Rule
Where:
CFc: the certainty factor of the consequent.
CFr: the certainty factor of the rule
CFf: the certainty factor of the fact

18. Semantic Context Management For
Ambiguous Data
4. Processing the ambiguity
– CF in FUZZY_CRISP Rules
Where S is the measure of similarity between both fuzzy sets and is calculated using
the following formula:
Where:
And:

19. Semantic Context Management For
Ambiguous Data
4. Processing the ambiguity
– Finally in the case of FUZZY_FUZZY Simple Rule the certainty factor of
the consequent is calculated using the same formula than in the
CRISP Simple RULE.
– Currently we do not support this type of combined reasoning for
complex rules that involve multiple clauses in their antecedent.

20. Future Work
• Create a mechanism that automatically assesses the certainty factor of a
sensor comparing its data with the one provided by other sensors.
– This will allow us to identify and discard malfunctioning sensors automatically.
• Develop an ecosystem of reasoners to distribute the inference process.
– We hope that this distribution will lead to a more agile and fast reasoning over the
context data, allowing us to combine less powerful devices to obtain a rich and
expressive inference
• Explore the possibility of including uncertainty in the membership
functions.

21. Thanks for your attention
Questions?
This work has been supported by project grant TIN2010-20510-C04-03
(TALIS+ENGINE), funded by the Spanish Ministerio de Ciencia e Innovación