Mars: a personalised mobile activity recognition system

Mobile activity recognition focuses on inferring the current activities of a mobile user by leveraging the sensory data that is available on today’s smart phones. The state of the art in mobile activity recognition uses traditional classification learning techniques. Thus, the learning process typically involves: i) collection of labelled sensory data that is transferred and collated in a centralised repository; ii) model building where the classification model is trained and tested using the collected data; iii) a model deployment stage where the learnt model is deployed on-board a mobile device for identifying activities based on new sensory data. In this paper, we demonstrate the Mobile Activity Recognition System (MARS) where for the first time the model is built and continuously updated on-board the mobile device itself using data stream mining. The advantages of the on-board approach are that it allows model personalisation and increased privacy as the data is not sent to any external site. Furthermore, when the user or its activity profile changes MARS enables promptly adaptation. MARS has been implemented on the Android platform to demonstrate that it can achieve accurate mobile activity recognition. Moreover, we can show in practise that MARS quickly adapts to user profile changes while at the same time being scalable and efficient in terms of consumption of the device resources.

Enhancement of a body area network to support smart health monitoring at the digital home

The deployment of home-based smart health services requires effective and reliable systems for personal and environmental data management. ooperation between Home Area Networks (HAN) and Body Area Networks (BAN) can provide smart systems with ad hoc reasoning information to support health care. This paper details the implementation of an architecture that integrates BAN, HAN and intelligent agents to manage physiological and environmental data to proactively detect risk situations at the digital home. The system monitors dynamic situations and timely adjusts its behavior to detect user risks concerning to health. Thus, this work provides a reasoning framework to infer appropriate solutions in cases of health risk episodes. Proposed smart health monitoring approach integrates complex reasoning according to home environment, user profile and physiological parameters defined by a scalable ontology. As a result, health care demands can be detected to activate adequate internal mechanisms and report public health services for requested actions.

Interacción desde dispositivos Android vía Bluetooth, con juguete teledirigido, para su uso por personas con discapacidad

El Proyecto Fin de Carrera, con título, "Interacción desde dispositivos Android vía Bluetooth, con juguete teledirigido, para su uso por personas con discapacidad" pretende completar la primera versión de la aplicación sobre dispositivo Android para manejar un juguete teledirigido añadiendo nuevas formas de interactuar con el dispositivo Android. Para este caso, el juguete teledirigido es el mismo: el robot esférico llamado Sphero. Dicho robot posee una interfaz a través de la cual puede recibir instrucciones, y mediante las cuales, se puede poner en movimiento o iluminarse con diferentes colores. Esta ampliación facilitará la interacción del usuario con el dispositivo Android, además de ampliar la funcionalidad hacia la inversa: recibir los movimientos del robot Sphero en el dispositivo Android cuando es manejado con la mano. Completando la primera versión, como cumplimiento de este Proyecto Fin de Carrera, se han realizado una serie de mandos nuevos, los cuales abarcan desde el manejo del robot por instrucciones de voz, movimientos del dispositivo Android desde el que se ejecuta, describir una trayectoria dibujada previamente en el dispositivo Android, o, realizar una serie de movimientos corregidos mediante la aplicación gracias a los límites de movimientos para un usuario concreto que pueden introducirse. Además, completando lo anteriormente escrito, se ha desarrollado una aplicación web que registrará todos los datos de uso del juguete, la cual, explota una base de datos que almacena toda interacción con el juguete. Estos datos estarán asociados a un usuario, debido a que la aplicación Android debe perfilarse para el uso de un usuario concreto. El perfilado de usuario se ha completado añadiendo una serie de información que puede ser útil para la persona que analice el comportamiento de una persona con discapacidad que utilice la aplicación. Por último, se ha realizado un estudio de elementos externos que puedan facilitar la interacción con los dispositivos Android a personas que sufran alguna discapacidad. ABSTRACT. The Thesis, titled "Interaction from Android devices via Bluetooth, with remote control toy, for use by people with disabilities" project aims to complete the first version of the application on Android device to manage a remote control toy adding new ways of interacting to Android device. For this case, the remote control toy is the same: the spherical robot called Sphero. This robot has an interface through which it can receive instructions, and by means of which can be set in motion or illuminated with different colors. This expansion will facilitate user interaction with the Android device, and extend the functionality to reverse: receiving Sphero robot movements in the Android device when handled by hand. Completing the first version, in fulfillment of this Thesis, there have been a number of new controls, which range from control of robot by voice instructions, movements Android device from which it runs, describe a path drawn previously on your Android device, or perform a series of movements corrected by applying through limits of movement for a particular user can be made. Besides completing the above written, has developed a web application that will record all data on use of the toy, which exploits a database that stores all interaction with the toy. These data will be associated with a user, because the Android application should be outlined for the use of a particular user. The user profile is completed by adding a range of information that can be useful for the person to analyze the behavior of a disabled person to use the application. Finally, a study was made of external elements that can facilitate interaction with Android devices to people who suffer from a disability.

Design and technical evaluation of an enhanced location-awareness service enabler for spatial disorientation management of elderly with mild cognitive impairment

The progressive ageing of population has turned the mild cognitive impairment (MCI) into a prevalent disease suffered by elderly. Consequently, the spatial disorientation has become a significant problem for older people and their caregivers. The ambient-assisted living applications are offering location-based services for empowering elderly to go outside and encouraging a greater independence. Therefore, this paper describes the design and technical evaluation of a location-awareness service enabler aimed at supporting and managing probable wandering situations of a person with MCI. Through the presence capabilities of the IP multimedia subsystem (IMS) architecture, the service will alert patient's contacts if a hazardous situation is detected depending on his location. Furthermore, information about the older person's security areas has been included in the user profile managed by IMS. In doing so, the service enabler introduced contribute to “context-awareness” paradigm allowing the adaptation and personalization of services depending on user's context and specific conditions or preferences.

Herramienta de partes de actividad

Para entender el proyecto de una manera muy simplificada lo podríamos resumir de la siguiente forma: Se trata de una aplicación que permite a los usuarios imputar las horas de las tareas que realizan de una forma sencilla, y que los gestores de estos usuarios, mediante informes ejecutivos, puedan analizar las dedicaciones y costes de los proyectos que éstos gestionan. Además se dispondrá de pantallas para administrar todas las entidades que están implicadas en las imputaciones. Las imputaciones se realizarán agrupándolas por el concepto “Parte”. Los Partes dividen los meses en grupos de 10 días, así cada usuario tendrá que imputar las horas que realiza cada 10 días, enviando a su supervisor el parte para que este pueda revisarlo y analizar los resultados de las imputaciones a través de los informes. Vamos a resumir las entidades que intervienen en la aplicación y sus relaciones de la siguiente forma: - Los usuarios tendrán asignados una serie de cargos, los cuales definirán las posibles tareas que podrán realizar categorizadas en los conceptos Proceso y Subproceso. Un ejemplo de esta relación podría ser el cargo “Programador” podrá realizar tareas de tipo Proceso “Desarrollo” y tipos de Subproceso “Programación” y “Pruebas unitarias” - También se asignarán a los usuarios los proyectos en los que trabajan. Estos proyectos pertenecerán a una unidad (se podría equiparar el concepto unidad por ejemplo a clientes, departamentos, etc). - También se podrán asignar a los usuarios las peticiones concretas de proyectos en los que intervienen. Una petición podría ser una incidencia detectada en un proyecto, una mejora concreta en un proyecto de mantenimiento, un soporte realizado para arreglar unos datos incorrectos, etc. - A los usuarios se le asignarán unos costes que podrán ser distintos por periodos de tiempo con el fin de conocer el coste exacto de una hora realizada en la fecha en la que fue realizada. - A los usuarios se les asignará un Perfil de usuario. Este perfil de usuario definirá los permisos sobre los módulos disponibles de la herramienta. Habrá un permiso por cada entidad, dividiéndolos en Alta, Baja, Modificación y Consulta. Además de estos permisos habrá permisos especiales sobre módulos que no se refieran al manteniendo de entidades pero que sí requieran de un permiso especial para poder ejecutarse como por ejemplo poder ejecutar los informes de costes, poder realizar y consultar partes de otros usuarios además de los propios del usuario, etc. Los perfiles serán dinámicos. Se podrán crear tantos perfiles como se requieran, cada uno con los permisos que se le asignen. - Los usuarios imputarán las horas que realizan categorizando por tarea los siguientes conceptos: Proyecto, Petición (opcional), Proceso, Subproceso, Descripción de la tarea y horas invertidas en los días de la decena. La aplicación permite poder adaptarse a las necesidades de diseño de la organización que la utilice y pueden crearse tantas instancias de la aplicación como se necesiten, compartiendo ono la base de datos entre varias de las instancias, sin necesidad de duplicar el código en el servidor, lo que facilita el mantenimiento de la aplicación cuando se añaden mejoras y se solucionan incidencias. También facilita el poder vender licencias de uso de la aplicación a otras entidades teniendo una sola versión de la aplicación en la nube, lo que hace muy bajo el coste de la venta de licencias aumentando así los beneficios por venta. ABSTRACT To understand the project in a very simplified way we could summarize as follows: Its an application that allows users to charge the hours of tasks that they realize in a easy way, and that the managers of these users, by executive reports, can analyze dedications and costs of projects that they manage. Furthermore screens will be available to manage all entities that are involved in the charge of hours. The charge of hours will be grouped by the concept "Timesheet". The Timesheets divided the months in groups of 10 days, so each user will have to charge the hours every 10 days by sending his supervisor the thimesheet so that it can review it and analyze the results of the charges of hours by reports.Let's summarize the entities involved in the application and their relationships as follows: - Users will be assigned to one or more positions, which define the possible tasks that can perform categorized into concepts Process and Subprocess. An example of this relationship could be the position "Developer" may do tasks of process type "Development" and subprocess types "Programming" and "Unit Testing" - Users will also be assigned to projects in which they work. These projects belongs to a unit (Some possible examples of unit could be customers, departments, ...). - You can also assign users to specific requests for projects thar they are involved. A request could be an issue detected in a project, a concrete improvement in a maintenance project, a support made to fix some incorrect data, ... - Users will be assigned to one or more costs that may be different for periods of time in order to know the exact cost of an hour on the date on which it was made. - Users will be assigned to a User Profile. This user profile define the available permissions on modules of tool. There will be a permission for each entity, dividing them into Insert, Delete, Modify and Read. In addition to these permissions will be special permissions on modules that are not related to maintaining entities but if require special permission to execute such as to execute the cost reports, to make and read Timesheets of others users in addition to themselve user, etc. The profiles will be dynamic. We can create many profiles as we required, each with the permissions that are assigned. - Users charge the hours that they work categorizing per task the following: Project, Request (optional), Process, Subprocess, task description and the hours that they used in the ten days period. The application allows to adapt to the needs of design of the organization that use the application and allows create many instances of the application as needed, yes or not sharing the database among multiple instances without duplicating the code on the server, which facilitates maintenance of the application when improvements are added and incidents are resolved. It also facilitates to sell licenses to use the application to other organizations having a single version of the application in the cloud, making it very low cost from the sale of licenses and increasing profits per sale.

AALUMO: a user model ontology for ambient assisted living services supported in next-generation networks

Ambient Assisted Living (AAL) services are emerging as context-awareness solutions to support elderly people?s autonomy. The context-aware paradigm makes applications more user-adaptive. In this way, context and user models expressed in ontologies are employed by applications to describe user and environment characteristics. The rapid advance of technology allows creating context server to relieve applications of context reasoning techniques. Specifically, the Next Generation Networks (NGN) provides by means of the presence service a framework to manage the current user's state as well as the user's profile information extracted from Internet and mobile context. This paper propose a user modeling ontology for AAL services which can be deployed in a NGN environment with the aim at adapting their functionalities to the elderly's context information and state.