**3. City4Age analytic framework**

These efforts increased research efforts in the area of smart city IoT data analytics through

Geriatric practice has in this aim utilized different standardized instruments based on traditional data collection methods (administration of questionnaires, meter-based measurement or direct observation in controlled conditions) and quantification and categorization of functional domains of daily life behavior and known frailty/MCI risk factors, such as Lawton IADL scale, Mini-Mental State Examination (MMSE), Fried Frailty Index, Nottingham Extended Activities of Daily Living, and numerous others. A comprehensive summary of such traditional generally **psychogeriatric instruments** and methods is provided in [3]. These instruments have evident major drawbacks, of late detection and problem identification (analysis and interpretation of questionnaires or conducting of exams can span intervals of months), and being generally ineffective, possibly subjective to a high degree, and costly for deployment.

The **City4Age project** (www.city4ageproject.eu), funded through the mentioned EC Horizon 2020 programme, is one of the pioneering efforts acting as a bridge between the mentioned two European Innovation Partnerships, EIP on SCC and EIP on AHA, contributing to specific and shared objectives and involving the committed participants from both Partnerships. The primary aim of the project is to enable fully Ambient Assisted age-friendly cities, through development and deployment of a range of ICT tools and services that will improve the unobtrusive early detection of MCI/frailty risks from heterogeneous IoT and smart city data sources at homes or on the move within the city, comprising the research and development work performed and results presented in this chapter as part of the work on the Data Analytics Platform. Coupled with the appropriate interventions—the developed tools will mitigate the detected risks as secondary aim. The developed system and components are being validated

Besides the City4Age project, there are numerous other related efforts in development of IoT driven systems for maintaining the functional independence and quality of life of the globally growing elderly population, or in development of data-driven health-related behavior recog-

The ActivAGE project (www.activageproject.eu), started in January 2017 and likewise funded through the Horizon 2020 programme, is a European multi-centric large-scale pilot on Smart Living Environments. The main objective is to build the first European IoT ecosystem across nine Deployment Sites (DS) in seven European countries, reusing and scaling up underlying open and proprietary IoT platforms, technologies and standards, and integrating new interfaces needed to provide interoperability across these heterogeneous platforms, that will enable the deployment and operation at large scale of active & healthy ageing IoT-based solu-

Participatory Urban Living for Sustainable Environments (PULSE) project (www.pulseproject.eu), started in January 2016 and likewise funded through the Horizon 2020 programme, harvests open city data, and data from health systems, urban and remote sensors, and personal devices, to enable evidence-driven and timely management of public health events and processes, leveraging diverse data sources and big data analytics to transform urban public health from a reactive to a predictive system, and from a system focused on surveillance to an

nition systems or platforms. Some of the recent relevant ones are the following:

through in-situ deployments in six pilot smart cities.

different projects.

118 Recent Applications in Data Clustering

tions and services.

Main challenges for the Data Science and Analytics in the related research in City4Age coincide with main generic challenges for the potential of collected IoT data from smart cities for health (and other personal) monitoring—volume and diversity of collected data is huge and promising, but the development of formalized and applicable knowledge and learning models is lagging with adequate potential for interpretation, classification and exploitation. At the same time, the unobtrusively acquired dataset for each specific person over time is often sparse, incomplete and erroneous, and with high degree of variation caused by temporary sensor imprecisions or influence of external factors beyond the sensing or modeling scope. High-level City4Age Analytics and detection process flow is depicted on **Figure 1**. City4Age has from the beginning adopted the combined hybrid knowledge- and data-driven approach, with the initial contribution of the knowledge-based approach turning out somewhat overestimated in the meantime due to the issues mentioned above and consequent non-deterministic and volatile semantic integrity of "known" or presumed universal geriatric causalities and concepts. The main focus is thus currently on the **data driven** behavior change variation recognition and characterization, analysis of relative changes in time series data for each specific person since the start of the pilot monitoring and determining baseline referent points, values and features (individual geriatric care analytics), and subsequently on discovering correlations and underlying features and interdependencies in the complete studied and monitored populations and clusters and groups within it (group exploratory analytics), starting from minimal initial domain model knowledge.

various sensing technologies and methods for collecting data (e.g., daily number of walked steps, weekly number of visits to relatives, daily time in seconds spent in public transport, etc.), as exemplified on the diagram above. Measures are analyzed and processed using various algorithmic techniques and/or methods, some of which are the clustering and grading algorithms described in this chapter, but others have also been tried and tested, and therein is the flexibility and scalability of the model and the Analytics Framework, supporting the registration of various algorithmic methods through metadata and deploying them on the "detection" variables (GFG, GEF, GES, Measures) on various model hierarchy and derivation levels. Example of network representation of GES, GEF and Measures is presented on the diagram on **Figure 2** below. Relations between nodes shown on the diagram are not fixed/ persistent, can variate according to different model configurations in different cities, or adap-

Temporal Clustering for Behavior Variation and Anomaly Detection from Data Acquired…

http://dx.doi.org/10.5772/intechopen.75203

121

The results of the data-driven detection are in turn used for expanding and building-up the domain knowledge base. The structure (ontologies and semantics) and mechanisms for this are established [9] are in parallel ongoing development, and will be still more intensely in the future work, in the scope and frame of City4Age contributions and breakthroughs in establishment of data-driven geriatrics. The unobtrusively acquired temporal dataset on individual level, currently being acquired for each single elderly person, is highly likely to expand with the increase and improvement of deployment scope and reliability of data acquisition

**Figure 2.** Example network representation of the City4Age geriatric model main constructs: Measures (purple), GES

tively according to the results of the data-driven detection.

and detection infrastructure and technologies.

(green), and GEF (red).

Majority of the functional domains and parameters of daily life behavior and geriatric risk are nevertheless known and established, and formalized in the City4Age hierarchic computational model of geriatric behavior and risk [8].

The main model constructs and variables are based on the notions of Geriatric factors (GEF), representing monthly behavior characterizations from all various functional behavioral domain variables and known MCI/frailty risk indicators, on unified Likert scale, with 1 denoting the least favorable and five the most favorable behavior with respect to MCI and Frailty risk, a common and standard adopted representation in geriatric practice and many of the used traditional instruments and questionnaires. GEF are further structured on several hierarchic levels of decomposition (GES—geriatric sub-factors, GFG—geriatric factor groups), and can be synthesized or derived from "Measures," native numeric values generated by the

**Figure 1.** High-level City4Age analytics and detection process flow.

various sensing technologies and methods for collecting data (e.g., daily number of walked steps, weekly number of visits to relatives, daily time in seconds spent in public transport, etc.), as exemplified on the diagram above. Measures are analyzed and processed using various algorithmic techniques and/or methods, some of which are the clustering and grading algorithms described in this chapter, but others have also been tried and tested, and therein is the flexibility and scalability of the model and the Analytics Framework, supporting the registration of various algorithmic methods through metadata and deploying them on the "detection" variables (GFG, GEF, GES, Measures) on various model hierarchy and derivation levels. Example of network representation of GES, GEF and Measures is presented on the diagram on **Figure 2** below. Relations between nodes shown on the diagram are not fixed/ persistent, can variate according to different model configurations in different cities, or adaptively according to the results of the data-driven detection.

promising, but the development of formalized and applicable knowledge and learning models is lagging with adequate potential for interpretation, classification and exploitation. At the same time, the unobtrusively acquired dataset for each specific person over time is often sparse, incomplete and erroneous, and with high degree of variation caused by temporary sensor imprecisions or influence of external factors beyond the sensing or modeling scope. High-level City4Age Analytics and detection process flow is depicted on **Figure 1**. City4Age has from the beginning adopted the combined hybrid knowledge- and data-driven approach, with the initial contribution of the knowledge-based approach turning out somewhat overestimated in the meantime due to the issues mentioned above and consequent non-deterministic and volatile semantic integrity of "known" or presumed universal geriatric causalities and concepts. The main focus is thus currently on the **data driven** behavior change variation recognition and characterization, analysis of relative changes in time series data for each specific person since the start of the pilot monitoring and determining baseline referent points, values and features (individual geriatric care analytics), and subsequently on discovering correlations and underlying features and interdependencies in the complete studied and monitored populations and clusters and groups within it (group exploratory analytics), starting from

Majority of the functional domains and parameters of daily life behavior and geriatric risk are nevertheless known and established, and formalized in the City4Age hierarchic computa-

The main model constructs and variables are based on the notions of Geriatric factors (GEF), representing monthly behavior characterizations from all various functional behavioral domain variables and known MCI/frailty risk indicators, on unified Likert scale, with 1 denoting the least favorable and five the most favorable behavior with respect to MCI and Frailty risk, a common and standard adopted representation in geriatric practice and many of the used traditional instruments and questionnaires. GEF are further structured on several hierarchic levels of decomposition (GES—geriatric sub-factors, GFG—geriatric factor groups), and can be synthesized or derived from "Measures," native numeric values generated by the

minimal initial domain model knowledge.

120 Recent Applications in Data Clustering

tional model of geriatric behavior and risk [8].

**Figure 1.** High-level City4Age analytics and detection process flow.

The results of the data-driven detection are in turn used for expanding and building-up the domain knowledge base. The structure (ontologies and semantics) and mechanisms for this are established [9] are in parallel ongoing development, and will be still more intensely in the future work, in the scope and frame of City4Age contributions and breakthroughs in establishment of data-driven geriatrics. The unobtrusively acquired temporal dataset on individual level, currently being acquired for each single elderly person, is highly likely to expand with the increase and improvement of deployment scope and reliability of data acquisition and detection infrastructure and technologies.

**Figure 2.** Example network representation of the City4Age geriatric model main constructs: Measures (purple), GES (green), and GEF (red).
