Page 145 - ITU KALEIDOSCOPE, ATLANTA 2019

AUTOMATIC PLAN GENERATING SYSTEM FOR GERIATRIC CARE
BASED ON MAPPING SIMILARITY AND GLOBAL OPTIMIZATION

2
1
Fei Ma ; Chengliang Wang ; Zhuo Zeng 3
1,2,3 Chongqing University, China

ABSTRACT smart home design, especially the difficulty in designing a
sensing devices selection plan (SDSP). Traditionally, in order
The smart home is an effective means of providing geriatric to implement a user-oriented smart home for geriatric care,
care to increase the ability of the elderly to live independently designers firstly collaborate with elderly and geriatric experts
and ensure their health in daily life. However, the smart to extract digital care demands based on natural language
home is not widely used because it is arduous to obtain a care needs. Then, designers choose corresponding services
sensing devices selection plan. In this paper, the accuracy to achieve care demands of the elderly based on their own
of service selection and cost savings assumes enormous comprehension, and select sensing devices to develop a SDSP
importance. Therefore, we propose an automatically plan by manually comparing the performance of sensing devices.
generating system for the elderly based on semantic similarity, Therefore, the traditional method limits the promotion and
intuitionistic fuzzy theory, and global optimization algorithm, use of the smart home in the field of geriatric care, including
aiming at searching for an optimized plan. Experiment results the following shortcomings and obstacles.
indicate that our approach can satisfy care demands and
provide an optimized plan of sensing devices selection. • Informal description: Since geriatric diagnosis and
smart services are deposited in the literature database
Keywords - Care demand, geriatric care, selection plan, without classification and arrangement, it increases the
smart home, smart service burden of designers to extract care demands of geriatric
diseases and select smart services.
1. INTRODUCTION
• Costly and casual design pattern: Due to the difficulty of
Due to low birth rates and low mortality rates and the manually extracting elderly demands, the elderly must
extension of life expectancy, the aging of population has pay expensive labor costs. Besides, it is casual and
been accelerating quickly and severely. An epidemiological arbitrary for designers to select services, since the elderly
study estimates that 11% of the world’s population is over do not understand the principles of smart services.
60 years old, but that figure is expected to rise to 22% by • Non-optimized plan: As it is arduous for designers
2050[1]. Accompanied by the substantial growth in the size to quantify the performance of sensing devices and
of the elderly population during the last several decades, to compare their performance manually, the traditional
the growing prevalence of geriatric diseases associated with SDSP is not optimized, and may result in surplus or
aging increases the burden on the health care systems. More inadequate sensing devices for implementing services.
importantly, geriatric diseases have a powerful negative
impact on perceived mental and physical functioning in In order to promote the use of the smart home in the
geriatric patients. It also increases life-threatening risks of the field of geriatric care, we have designed an Automatic Plan
elderly[2], for example, heart disease and stroke account for Generating System (APGS) that automatically generates the
more than 40% of all deaths among persons aged 65-74 and optimized SDSP. The contributions of this paper are given
almost 60% of those aged 85 years[3]. Therefore, geriatric below.
care plays an important role in maintaining good health and • UDSD architecture: We proposed a UDSD (user
increasing the life quality of the elderly. demand service device) architecture, which uses key
Recently, with remarkable advancements in machine learning labels instead of natural language descriptions to
and artificial intelligence, the smart home emerges into formalize user information, care demands, smart
the public consciousness, because of its convenience and services, and sensing devices, including the user layer,
precision in health care[4][5][6]. Smart homes could not demand layer, service layer and device layer.
only monitor the daily living of the elderly and assist living
in patients, but also evaluate the emotions of inhabitant • Smart-desire mapping method: Due to the high cost
and warn of dangers[7][8][9]. However, the smart home and casual selection of traditional design method, we
is not widely used in the field of geriatric care, because of designed a Smart-desire mapping method (SDMM)
two main reasons: (1) the challenge of extracting required to automatically extract atomic care demands based
smart services for elderly with high precision, based on on expert knowledge and mapping smart services by
user demands; 2) the huge complexity and high costs of calculating semantic similarity and QoS similarity.

978-92-61-28401-5/CFP1968P-ART @ ITU 2019 – 125 – Kaleidoscope