DIVIDE: ADAPTIVE CONTEXT-AWARE QUERY DERIVATION FOR IOT DATA STREAMS - PowerPoint PPT Presentation

DIVIDE: ADAPTIVE CONTEXT-AWARE QUERY DERIVATION FOR IOT DATA STREAMS Mathias De Brouwer Sensors and Actuators on the Web (SAW) workshop – ISWC 2019 Auckland, New Zealand – 26 October 2019

CONTEXT-AWARE MONITORING IN I O T Wearable Motion sensor Sound sensor A0 Light sensor A1 Temperature sensor A2 … Localization Door/window sensor 3

CONTEXT-AWARE MONITORING IN I O T Wearable Motion sensor Sound sensor A0 Light sensor A1 Temperature sensor A2 … Localization Door/window sensor Medical domain Hospital lay-out Care staff Electronic Health knowledge 4 Record of patients

CONTEXT-AWARE MONITORING IN I O T Wearable Motion sensor Sound sensor A0 Light sensor A1 Temperature sensor A2 … ? Localization Door/window sensor Medical domain Hospital lay-out Care staff Electronic Health knowledge 5 Record of patients

CONTEXT-AWARE MONITORING IN I O T Wearable Motion sensor Sound sensor A0 R EQUIRES INTELLIGENT Light sensor A1 Temperature sensor A2 CONSOLIDATION & ANALYSIS … OF HETEROGENEOUS , VOLUMINOUS , HIGH VELOCITY DATA ? Localization Door/window sensor Medical domain Hospital lay-out Care staff Electronic Health knowledge 6 Record of patients

CONTEXT-AWARE MONITORING IN I O T Wearable Motion sensor Sound sensor A0 R EQUIRES INTELLIGENT Light sensor A1 Temperature sensor A2 CONSOLIDATION & ANALYSIS … OF HETEROGENEOUS , VOLUMINOUS , HIGH VELOCITY DATA RDF STREAM Localization PROCESSING Door/window sensor Medical domain Hospital lay-out Care staff Electronic Health knowledge 7 Record of patients

EXAMPLE USE CASE – PATIENT BOB Wearable Motion sensor Sound sensor A0 Light sensor A1 Temperature sensor A2 … RDF STREAM Localization PROCESSING Door/window sensor Medical domain Hospital lay-out Care staff Electronic Health knowledge 8 Record of patients

EXAMPLE USE CASE – PATIENT BOB Wearable Motion sensor Sound sensor A0 Light sensor A1 Temperature sensor A2 … RDF STREAM Localization PROCESSING Door/window sensor Diagnosis Bob: epilepsy attack Medical domain Hospital lay-out Care staff Electronic Health knowledge 9 Record of patients

EXAMPLE USE CASE – PATIENT BOB Wearable Motion sensor Sound sensor A0 Light sensor A1 Temperature sensor A2 … RDF STREAM Localization PROCESSING Door/window sensor Diagnosis Bob: Epilepsy patients are sensitive to light epilepsy attack Threshold: light should not go > 250 lumen Medical domain Hospital lay-out Care staff Electronic Health knowledge 10 Record of patients

EXAMPLE USE CASE – PATIENT BOB Wearable Motion sensor Sound sensor A0 Query 1: generate Light sensor A1 alarm if value Temperature sensor A2 measured by sensor … A1 is > 250 RDF STREAM Localization PROCESSING Door/window sensor Diagnosis Bob: Epilepsy patients are sensitive to light epilepsy attack Threshold: light should not go > 250 lumen Medical domain Hospital lay-out Care staff Electronic Health knowledge 11 Record of patients

EXAMPLE USE CASE – PATIENT BOB Wearable Motion sensor Sound sensor A0 Query 1: generate Light sensor A1 alarm if value Temperature sensor A2 measured by sensor … A1 is > 250 RDF STREAM Localization PROCESSING Door/window sensor Diagnosis Bob: concussion Medical domain Hospital lay-out Care staff Electronic Health knowledge 12 Record of patients

EXAMPLE USE CASE – PATIENT BOB Wearable Motion sensor Sound sensor A0 Query 1: generate Light sensor A1 alarm if value Temperature sensor A2 measured by sensor … A1 is > 250 RDF STREAM Localization PROCESSING Door/window sensor Concussion patients are sensitive to light & sound Diagnosis Bob: Thresholds: light should not go > 170 lumen, concussion sound should not go > 30 decibels Medical domain Hospital lay-out Care staff Electronic Health knowledge 13 Record of patients

EXAMPLE USE CASE – PATIENT BOB Wearable Motion sensor Sound sensor A0 Query 1: generate Light sensor A1 alarm if value Temperature sensor A2 measured by sensor … A1 is > 250 170 RDF STREAM Localization PROCESSING Door/window sensor Concussion patients are sensitive to light & sound Diagnosis Bob: Thresholds: light should not go > 170 lumen, concussion sound should not go > 30 decibels Medical domain Hospital lay-out Care staff Electronic Health knowledge 14 Record of patients

EXAMPLE USE CASE – PATIENT BOB Wearable Motion sensor Sound sensor A0 Query 1: generate Query 2: generate Light sensor A1 alarm if value alarm if value Temperature sensor A2 measured by sensor measured by sensor … A1 is > 250 A0 is > 30 170 RDF STREAM Localization PROCESSING Door/window sensor Concussion patients are sensitive to light & sound Diagnosis Bob: Thresholds: light should not go > 170 lumen, concussion sound should not go > 30 decibels Medical domain Hospital lay-out Care staff Electronic Health knowledge 15 Record of patients

HOW TO DEAL WITH CONTEXT CHANGES? Real-time How to Manual reconfiguration Approach reasoning? configure? when context changes? F IXED GENERIC expressive manually no QUERIES 16

HOW TO DEAL WITH CONTEXT CHANGES? Real-time How to Manual reconfiguration Approach reasoning? configure? when context changes? F IXED GENERIC expressive manually no QUERIES 17

HOW TO DEAL WITH CONTEXT CHANGES? Real-time How to Manual reconfiguration Approach reasoning? configure? when context changes? F IXED GENERIC expressive manually no QUERIES 18

HOW TO DEAL WITH CONTEXT CHANGES? Real-time How to Manual reconfiguration Approach reasoning? configure? when context changes? F IXED GENERIC expressive manually no QUERIES M ULTIPLE QUERIES (almost) (1 FOR EACH manually yes none RELEVANT SENSOR ) 19

HOW TO DEAL WITH CONTEXT CHANGES? Real-time How to Manual reconfiguration Approach reasoning? configure? when context changes? F IXED GENERIC expressive manually no QUERIES M ULTIPLE QUERIES (almost) (1 FOR EACH manually yes none RELEVANT SENSOR ) 20

HOW TO DEAL WITH CONTEXT CHANGES? Real-time How to Manual reconfiguration Approach reasoning? configure? when context changes? F IXED GENERIC expressive manually no QUERIES M ULTIPLE QUERIES (almost) (1 FOR EACH manually yes none RELEVANT SENSOR ) 21

HOW TO DEAL WITH CONTEXT CHANGES? Real-time How to Manual reconfiguration Approach reasoning? configure? when context changes? F IXED GENERIC expressive manually no QUERIES M ULTIPLE QUERIES (almost) (1 FOR EACH manually yes none RELEVANT SENSOR ) DIVIDE none automatically no 22

GENERAL ISSUE VS. GOAL (Changed) context 23

GENERAL ISSUE VS. GOAL Queries of interest are (Changed) automatically derived & context (re)configured 24

GENERAL ISSUE VS. GOAL Queries of interest are (Changed) automatically derived & context (re)configured By performing reasoning on (changed) context instead of on the real-time data streams 25

GENERAL ISSUE VS. GOAL DIVIDE Queries of interest are (Changed) automatically derived & context (re)configured By performing reasoning on (changed) context instead of on the real-time data streams 26

DIVIDE – BUILDING BLOCKS Logic: Notation3 Logic (N3) 27

DIVIDE – BUILDING BLOCKS Logic: Notation3 Logic (N3) Reasoner: EYE reasoner ➔ OWL 2 RL reasoning ➔ define goal (= which triples EYE should look for evidence) ➔ EYE produces a proof with the goal as the last applied rule 28

OVERVIEW OF DIVIDE New or changed (e.g., for a patient or room) ONTOLOGY O NTOLOGY C ONTEXT PREPROCESSING Output: filtering queries to run on RSP engine P REPROCESSED O NTOLOGY ( IN EYE IMAGE ) QUERY I NSTANTIATED DERIVATION Q UERY S ENSOR Q UERY R ULE R EASONER ’ S G OAL 29

ONTOLOGY PREPROCESSING Goal: speed up query derivation 30

ONTOLOGY PREPROCESSING Goal: speed up query derivation How? 31

ONTOLOGY PREPROCESSING Goal: speed up query derivation How? Create N3 copy of ontology 32

ONTOLOGY PREPROCESSING Goal: speed up query derivation How? Create specialized ontology-specific Create N3 copy of ontology rules from OWL 2 RL rules 33

ONTOLOGY PREPROCESSING Goal: speed up query derivation How? Create specialized Create compiled ontology-specific prolog image of Create N3 copy of ontology rules from OWL 2 EYE (with ontology RL rules & specialized rules) 34

DIVIDE – BUILDING BLOCKS Ontology: ACCIO ontology 35

DIVIDE – BUILDING BLOCKS Ontology: ACCIO ontology LightIntensityAboveThresholdFault ≡ (hasSymptom some LightIntensityAboveThresholdSymptom) and (madeBySensor some (isSubsystemOf some (hasLocation some (isLocationOf some ( (hasDiagnosis some (hasMedicalSymptom some SensitiveToLight)) and (hasRole some PatientRole)))))) 36

DIVIDE – BUILDING BLOCKS Ontology: ACCIO ontology HandleHighLightInRoomAction ≡ LightIntensityAboveThresholdFault and (madeBySensor some (isSubsystemOf some (hasLocation some (isLocationOf some LightingDevice)))) 37

INPUTS OF QUERY DERIVATION Reasoner goal : look for an AboveThresholdAction {?x a :AboveThresholdAction.} => {?x a : AboveThresholdAction.}. 38