Exploring Digital Technologies for Addressing Risk Factors of Solitary Death in South Korea

Abstract

South Korea has experienced rapid aging, and it is predicted that people aged 65 years or older will account for 20% of the total population by 2047. The proportion of one-person households is also increasing rapidly at the same time, and is expected to account for 37% of the total number of households by 2047. Such a demographic shift has led to more isolated households and social isolation, and consequently, to higher risks of solitary deaths. Central and local governments have implemented various measures, such as care services for the elderly living alone and regular checks on their status; however, there are limitations due to workforce and budget constraints. Therefore, it is necessary to explore innovative solutions that utilize digital technologies such as the internet of things (IoT) and artificial intelligence (AI) as new low-cost, high-efficiency alternatives. Innovations in digital technology have the potential to suggest new solutions to the issue of solitary death. Advanced technologies can play an important role in maximizing the effectiveness of solitary death prevention, management, and the prevention of social isolation. Although there is a growing body of research on the development of digital technology-based care services for the elderly, these studies have primarily focused on the applicability and early implementation stages of specific technologies. These studies have limitations in fully understanding the market trends of technologies and competition patterns within the industry. Therefore, it is essential to investigate trends in the latest technologies and R&D directions in a particular technology field through patents filed to protect new technologies. This study is intended to contribute to an in-depth understanding of the current status and development direction of digital technologies for the prevention of solitary deaths, and to provide a basis for future technology development. This study also contributes to establishing government policies and R&D strategies.

1. Introduction

Solitary death has recently become a serious social issue owing to the aging population and the increasing number of one-person households. Solitary death refers to living and dying alone, a tragic situation that undermines human dignity. Statistics have revealed that the number of solitary deaths has gradually increased, with thousands of cases recorded annually. In this sense, solitary death has been recognized as a social issue and not just an individual problem. South Korea has experienced rapid aging, and it is predicted that people aged 65 years or older will account for 20% of the total population. The proportion of one-person households is also increasing rapidly at the same time, and is expected to account for 37% of the total number of households by 2047. Such a demographic shift has led to more isolated households and social isolation, and consequently, to higher risks of solitary deaths. Central and local governments have implemented various measures, such as care services for the elderly living alone and regular checks on their status; however, there are limitations due to workforce and budget constraints. Therefore, it is necessary to explore innovative solutions that utilize digital technologies such as the internet of things (IoT) and artificial intelligence (AI) as new low-cost, high-efficiency alternatives.

Innovations in digital technology have the potential to suggest new solutions to the issue of solitary death. IoT enables real-time monitoring of the daily activities of the elderly living alone and immediately alters in the event of anomalies occurring; AI predicts health conditions and psychological changes by analyzing collected data. Additionally, the analysis of big data on patterns and factors related to solitary deaths in communities can be utilized to design policies and programs. These advanced technologies can play an important role in maximizing the effectiveness of solitary death prevention, management, and the prevention of social isolation.

Although there is a growing body of research on the development of digital technology-based care services for the elderly [1,2,3,4]; however, studies have primarily focused on the applicability and early implementation stages of specific technologies. Thus, these studies have limitations in fully understanding the market trends of technologies and competition patterns within the industry. Therefore, it is essential to investigate trends in the latest technologies and R&D directions in a particular technology field through patents filed to protect new technologies.

This study aims to explore the use of digital technologies in depth to address the risk factors of solitary deaths with the following specific objectives.

First, it identifies the annual trend of patent applications in South Korea related to solitary deaths. Second, it identifies the current status of patent classifications in South Korea and explores their specific characteristics. Third, it examines key digital technologies for preventing solitary deaths. Fourth, it clarifies the main technology areas for responding to solitary deaths by identifying closely related digital technologies.

Based on these objectives, this study intends to contribute to an in-depth understanding of the current status and development direction of digital technologies for the prevention of solitary deaths, and to provide a basis for future technology development. This study also contributes to establishing government policies and R&D strategies. Furthermore, this study provides a foundation for the development of convergence technologies to address the issue of solitary deaths.

2. Theoretical Background

2.1. Digital Technologies for Solitary Death Prevention

Central and local governments and private companies have distributed small devices with various features under pilot projects to monitor the health status of the elderly living alone, support residential safety and convenience, and provide emotional support and information. Local governments have been actively distributing devices with comprehensive smart home service functions to the elderly living alone. Digital technologies for solitary death prevention include wearable devices, smart homes, mobile applications, AI, big data, virtual reality, and augmented reality. Previous studies have applied them for prevention.

Among the studies on the application of wearable devices, the function of real-time monitoring of data on lifestyle patterns and health status of the elderly living alone through smart wearable bands has been emphasized [5]. Previous studies have analyzed the recent trends in the development of wearable healthcare devices for the elderly and explored the balance between technological advancements and user convenience [3], and studied the implementation of wearable devices that can immediately detect health risks in the elderly in real time and respond to emergencies [6].

Other smart home-related studies have not focused on solitary deaths but have suggested various means to support living safety and emotional health by applying relevant technologies to the daily lives of vulnerable groups such as the elderly and the disabled. Lee et al. [7] proposed the prevention of depression in the elderly through a mobile platform using a smart home garden, and Bae [8] analyzed the case of a smart home-based living lab in South Korea to facilitate domiciliary care and welfare technology in the new normal era for the disabled and elderly. Koo et al. [9] developed a smart home CCTV system that enhanced user safety via object recognition.

Studies have applied mobile technologies to strengthen the safety and emotional support of the elderly living alone. Song and Moon [10] developed an emergency safety and security service using mobile technologies to enable the elderly living alone to quickly ask for help in emergency situations, thus strengthening the safety of the elderly living alone and reducing the risks of solitary deaths. Bum et al. [11] analyzed the effects of mobile media technologies on alleviating suicidal thoughts in the elderly and examined the potential for technologies to strengthen social connectedness and support mental health. Yoo et al. [12] developed a system to monitor the water consumption of the elderly living alone via a smartphone application and take appropriate measures when an abnormality is detected, suggesting a way to manage the daily life patterns of the elderly living alone and improve their safety.

Additionally, there are other studies that have explored innovative technological approaches to lower the risk of solitary deaths by applying digital technologies such as IoT and big data. Lee [13] explained how to construct a system to collect the health data of the elderly living alone in real time through diverse IoT devices and send immediate notifications when abnormal activities are detected. Lim and Lee [14] collected and analyzed daily data of the elderly living alone using smart home technologies, wearable devices, and mobile apps. They also studied measures to detect health abnormalities or risk conditions early and provide necessary medical intervention or social support. Heo and Park [15] proposed a system that utilizes IoT sensors to monitor heart rates, activity levels, and sleep patterns, continuously tracking the health status of the elderly and swiftly responding to emergency situations.

2.2. Patent Information-Based Research

Patents provide patent application information generated from the results of technological inventions through certain procedures [16], representing 80% of knowledge [17]. Patent information is highly useful data that enable the exploration of R&D activities, the level of technological development, and new technologies in the future. They include the bibliography of an invention, such as its name, inventor, application date, international patent classification, and citations, and provide quantitative and objective information that can be analyzed from diverse perspectives [18].

The International Patent Classification (IPC) code is a classification system announced by the World Intellectual Property Organization and is a patent classification system for systematizing the classification, search, distribution, and management of patent information [19]. The IPC has a system of section, class, subclass, main group, and subgroup. When an IPC code is assigned, only one main classification code is given if the invention has a single technical content (including function and use). However, if there are several technical contents, the most representative technical content of the applied invention is designated as the “main classification” and the remaining technical contents are categorized as “subclassification” [20].

Patent analysis is a research method that selects important keywords from officially registered Korean and non-Korean patent documents and identifies technological development trends and potentially promising technology areas based on the relationships between the selected keywords.

Recently, patents have been analyzed using various big data analysis techniques. In South Korea, patent analyses have been conducted on new technologies in the sports industry [21], the solar power technology patents of small and medium enterprise [22], fusion technology in the fishery sector [23], radioactive fusion technology [24]. In other countries, patent analyses have been conducted on wastewater chemical treatment technologies [25], AI chatbot technologies [26], and cybersecurity [27]. These studies present the applicability of patent data for understanding technological trends and associations in various technology fields.

3. Research Method

3.1. Research Stage

Table 1. Research stages and analysis tools.

Steps		Analysis Details	Analysis Tools
Data collection		Construction of solitary death-related text data	KIPRIS
Data preprocessing		Elimination of postpositions and special letters, as well as keyword unification	TEXTOM
Frequency analysis		Analysis of IPC codes or keyword frequencies
Social network analysis	Centrality analysis	Calculation of network centrality	UCINET 6.721, NetDraw 2.176
	Structural equivalence analysis: CONCOR	Keyword grouping/clustering

presents the stages of the study. First, we searched and collected patent data related to solitary death prevention from the KIPRIS database. The search keywords encompassing words similar to “solitary death, digital technology”, and highly relevant patent data were selected. We also refined the collected patent data for easier analysis. We extracted the IPC codes and summary information for each patent to construct the dataset. Data accuracy improved after removing unnecessary information and duplicate data. The number of patents per year and the main technologies were confirmed through frequency analysis. To identify the connectivity and structural relationships of the technologies, we constructed a network and defined the digital technologies in each patent as nodes based on the refined data. A centrality analysis was performed to evaluate the importance of each node (digital technology) in the network. Such an analysis enabled the identification of key digital technologies that play an important role in preventing solitary deaths. A structural equivalence analysis, specifically the convergence of iterated correlations (CONCOR), was performed to cluster technologies that play similar roles within the network structure. This attempt enabled the identification of patterns of technologies with similar statuses and functions within the network.

3.2. Research Targets and Data Collection

This study targeted the full texts of patents filed for solitary death prevention. Data were collected according to the conditions in

Table 2. Data collection procedure for digital technology patents on solitary death prevention.

Category	Description
Collection period	January 2005~June 2024
Status	Application and disclosure
Intellectual property rights	Domestic patents
Search keywords	“solitary death”, “lonely death”, “dying alone”, “unattended death”, + “digital”, “technology”, “system”, “deep learning”, “AI”, “big data”, “IoT”, etc.

. We utilized KIPRIS, a patent information search service provided by the Korean Intellectual Property Office, to collect data. We searched for patents filed or released by June 2024 starting from the date when the data were searchable on KIPRIS. Among intellectual property rights, only patents in South Korea were targeted; utility model patents, foreign patents, trademarks, and designs were excluded from the analysis.

The main search keywords were selected based on expressions commonly used in various literature and articles addressing the phenomenon of solitary death. The term “solitary death” originated in Japan in the 1990s as a neologism following a significant increase in deaths occurring alone, gaining broader attention after a 2011 broadcast and later being referred to as “lonely death” in a 2022 CNN article. Drawing on this background, this study selected keywords such as “solitary death”, “lonely death”, “dying alone”, and “unattended death” to encompass the various expressions related to solitary death. Similar keywords were chosen to comprehensively reflect key areas of digital technologies for preventing solitary death and incorporate the latest technological trends. Initially, using the main search keywords alone, a total of 1432 patents were retrieved. Next, to closely examine patents related to digital technologies for the prevention of solitary deaths, additional searches were conducted by combining digital technology-related keywords. Keywords such as “digital”, “technology”, “system”, “deep learning”, “AI”, “big data”, and “IoT” were combined with the main search keywords, resulting in a total of 1273 patents being retrieved. It was found that approximately 88.9% of the patents related to solitary death were developed based on digital technologies. Among these patents, the titles and abstracts were reviewed to identify those that directly addressed technologies related to the prevention of solitary deaths. Ultimately, 813 patents directly related to the prevention of solitary deaths were selected.

3.3. Data Preprocessing

We preprocessed the data for analysis and removed parentheses, punctuation marks, and stop words using TEXTOM. The search keywords for research, such as “solitary death”, “living alone”, and “solitary death”, were deleted, and the terms were unified by changing the terms in English to those in Korean or correcting the spacing to extract technology-related keywords from the collected data. Partial examples of the modified and unified words are listed in

Table 3. Patent data preprocessing steps and methods.

Collected Words	Modified Words
AI, artificial intelligence	AI
VR, virtual reality	VR
AR, augmented reality	AR
IoT, internet of things	IoT

.

3.4. Frequency Analysis

We confirmed the annual trend of the collected patent data through frequency analysis, as well as the current status of the main and subclassifications of IPC codes. Specifically, we selected the top 50 keywords that are directly and indirectly related to digital technologies, except for keywords indicating the functions of digital technologies, such as “management”, “life”, and “collection”.

3.5. Social Network Analysis: Centrality Analysis

Social network analysis consists of nodes indicating various relationships and links presenting a connection status based on data, such as individuals, groups, and knowledge [1]. The importance of nodes varies within the network, depending on their size, and links are lines that represent the relationships between nodes. Degree centrality is an indicator that measures how much a particular word is connected to other words within the network and is utilized to indicate the word’s positional dominance, importance, and influence [28]. This indicator intuitively indicates the importance of a particular word, and a high degree centrality indicates that the word has a high influence on the network [29]. This study utilized the degree centrality shown in

Table 4. Explanation of the centrality algorithm.

Category	Description
Degree Centrality	Degree of connection with other nodes
Closeness Centrality	Shortest path distance from another node
Betweenness Centrality	The extent to which it is located at the shortest distance in the network
Eigenvector Centrality	Degree of connection with important nodes in the network

and confirmed the complex network of digital technologies for solitary death prevention.

3.6. Structural Equivalence Analysis: CONCOR

CONCOR, a type of structural equivalence analysis, is an indicator that clusters words with similar positions based on their correlation coefficients, analyzes the relationships between clusters, and evaluates the importance of each word [30]. It focuses on analyzing the connection patterns between structurally similar words rather than the direct or indirect connections across the entire network [31]. This study confirmed the clusters and structural features of digital technologies for solitary death prevention through structural equivalence analysis.

4. Research Results

4.1. Year-to-Year Trend of Patent Applications for Solitary Death Prevention

Table 5. Year-to-year trends in digital technology patent applications for solitary death prevention.

Year	Frequency	Percentage	Year	Frequency	Percentage
2005	2	0.2	2015	29	3.6
2006	4	0.5	2016	36	4.4
2007	2	0.2	2017	48	5.9
2008	8	1.0	2018	59	7.3
2009	8	1.0	2019	90	11.1
2010	9	1.1	2020	102	12.5
2011	5	0.6	2021	150	18.5
2012	17	2.1	2022	160	19.7
2013	26	3.2	2023	41	5.0
2014	17	2.1	Total	813	100.0

and Figure 1 show the results of confirming the annual trends of a total of 813 patents in South Korea related to the prevention of lonely deaths. Since its first application in 2005, it has been in use until 2023. A high number of applications was found in 2022 (160), 2021 (150), and 2020 (102) with an increasing trend beginning in 2015. On the other hand, there was a sharp decline in patent applications in 2023 (41). This may be attributed to the fact that the COVID-19 pandemic, which began in 2019, increased social isolation, leading to a surge in research and technological development for the prevention of solitary deaths. However, as the COVID-19 situation stabilized in 2023, research and development activities may have slowed down. Another possible reason for the decline could be the time lag between patent application and publication. Typically, patents are not disclosed until 18 months after they are filed, so many of the patents filed in 2023 may not have been disclosed yet and therefore were not captured in the data.

4.2. Patent Analysis Based on International Patent Classification (IPC)

IPC is an internationally unified patent classification system that indicates technology areas of inventions and consists of sections, classes, subclasses, main groups, and subgroups. The highest level of the IPC classification (section) is categorized into eight sections from A to H: daily necessities under A, processing manipulation and transportation under B, chemistry and metallurgy under C, textile and paper under D, fixed structures under E, mechanical engineering, lighting, heating, weapons, and explosion under F, physics under G, and electricity under H.

This study targeted and analyzed all international patent classifications of technology patents related to solitary death prevention filed in KIPRIS. As shown in

Table 6. Patent analysis based on international patent classification (IPC) for solitary death prevention technologies.

Section	Description	Frequency	Percentage
A	Daily necessities	681	17.0
B	Processing manipulation and transportation	117	2.9
C	Chemistry and metallurgy	44	1.1
D	Textile and paper	12	0.3
E	Fixed structures	24	0.6
F	Mechanical engineering, lighting, heating, weapons, and explosions	68	1.7
G	Physics	2557	63.7
H	Electricity	510	12.7
Total		4013	100

, the section with the most applications was physics (63.7%), followed by daily necessities (17%), electricity (12.7%), and processing/transportation (2.9%).

There were 49 patents for single technologies registered as IPC main classification codes, while 764 patents were registered as IPC main classification and subclassification codes.

4.3. IPC Main Classification

We examined the IPC main classification of the patents to identify specific digital technology patents. Among the 813 patents, G08B 21/04 (response to inactivity) accounted for the highest proportion, followed by G06Q 50/22 (social services or social welfare), and G06Q 50/10 (supply, production, management, and editing of content via services and communication networks). Eight of the top 10 main classifications were found in Section G, whereas A61B 5/00 (measurement for diagnosis and personal identification) and H04L 12/28 (featured by path configuration) accounted for the highest proportions. The top 10 IPC main classification codes are listed in

Table 7. IPC main classifications for solitary death prevention technologies.

Rank	IPC	Description	Frequency
1	G08B 21/04	Response to inactivity, e.g., inactivity of the elderly	156
2	G06Q 50/22	Social services or social welfare, e.g., community support activities or consultation services	100
3	G06Q 50/10	Supply, production, management, and editing of content via services and communication networks	46
4	A61B 5/00	Measurement for diagnosis and personal identification	39
5	G08B 21/02	Alert to ensure personal safety	22
6	G16H 50/20	For diagnosis by computer use, e.g., based on the medical expert system	18
7	G06Q 50/26	Government or public service (business processes related to the transportation industry)	17
8	H04L 12/28	Featured by path configuration, e.g., wireless communication network	11
9	G16H 20/70	Psychotherapy-related content, e.g., psychological therapy or self-training	10
10	G16H 50/30	For the calculation of health indicators; for the assessment of individual health risks	10

.

4.4. IPC Subclassification

There were 3189 subclassifications of the 813 patents, of which the top five IPC codes were the same as the IPC main classification: G08B 21/04 (response to inactivity), G06Q 50/22 (social services or social welfare), G06Q 50/10 (supply, production, management, and editing of content via services and communication networks), A61B 5/00 (measurement for diagnosis and personal identification), and G08B 21/02 (alert to ensure personal safety). Regarding subclassification, G08B 25/14 (central alert receiver or notification device), G08B 25/10 (using a wireless transmission system), and G08B 21/18 (for patient-specific data) showed high frequencies, unlike the main classification. The top 10 IPC subclassification codes are listed in

Table 8. IPC subclassifications for solitary death prevention technologies.

Rank	IPC	Description	Frequency
1	G08B 21/04	Response to inactivity, e.g., inactivity of the elderly	294
2	G06Q 50/22	Social services or social welfare, e.g., community support activities or consultation services	150
3	G06Q 50/10	Supply, production, management, and editing of content via services and communication networks	146
4	A61B 5/00	Measurement for diagnosis and personal identification	122
5	G08B 21/02	Alert to ensure personal safety	111
6	G08B 25/14	Central alert receiver or notification device	100
7	G08B 25/10	Using a wireless transmission system	75
8	G08B 21/18	Status alert	56
9	G16H 10/60	For patient-specific data, e.g., for electronic patient records	55
10	G06Q 50/26	Government or public service (business processes related to the transportation industry)	54

.

4.5. Subclassifications of G08B 21/04

As a result of examining the main classification, when G08B 21/04 (response to inactivity), which is the most frequently filed IPC, is the main classification, G08B 25/14 (central alert receiver or notification device) was found to be the most frequently appearing code as the subclassification. Subsequently, communication and alarm technologies converged: G08B 25/10 (using a wireless transmission system), G08B 21/02 (alert to ensure personal safety), and G08B 21/18 (status alert). Additionally, there was convergence with G06N 20/00 (machine learning) connected to social services, such as G06Q 50/22 (social services or social welfare) and G06Q 50/26 (government or public service). Subclassifications codes for the main category, G08B 21/04, are shown in

Table 9. Frequency analysis of subclassifications in G08B 21/04.

Rank	IPC Code	Description	Frequency
1	G08B 25/14	Central alert receiver or notification device	48
2	G08B 25/10	Using a wireless transmission system	36
3	G08B 21/02	Alert to ensure personal safety	34
4	G08B 21/18	Status alert	16
5	G08B 3/10	Using electricity transmission modes	14
7	G06Q 50/22	Social services or social welfare	13
8	G06N 20/00	Machine learning	11
9	G06Q 50/10	Services such as communication network, internet, web, monitoring control, etc.	9
	G08B 25/01	What is featured by the transmission medium	9
	G06N 3/08	Learning method	9
10	G06Q 50/26	Government or public service	8

.

4.6. Subclassifications of G06Q 50/22

When G06Q 50/22 (social services or social welfare), which is the second most frequently filed IPC, was the main classification, the most frequently appearing subclassification was G08B21/04 (response to inactivity). Subsequently, there was convergence with G06Q50/10 (services such as communication networks, internet, web, monitoring control, etc.), G06Q50/26 (government or public service), and H04W4/80 (services using short-range communication). It also converged with medical technology, such as G16H10/60 (patient-specific data) and A61B5/00 (measurement for diagnosis). Subclassifications codes for the main category, G06Q 50/22, are shown in

Table 10. Frequency analysis of subclassifications in G06Q 50/22.

Rank	IPC Coded	Description	Frequency
1	G08B21/04	Response to inactivity, e.g., inactivity of the elderly	55
2	G06Q50/10	Services such as communication network, internet, web, monitoring control, etc.	48
3	G06Q50/26	Government or public service	17
4	H04W4/80	Services using short-range communication, e.g., near-field communication	13
5	H04W4/02	Services using location information of users or mobile device/terminal	12
6	G16H10/60	For patient-specific data, e.g., for electronic patient records	11
7	G08B25/00	Alarm system to notify the central office of the location of the alarm status, e.g., fire or police telegraph system	10
8	G08B21/02	Alert to ensure personal safety	10
9	A61B5/00	Measurement for diagnosis	10
10	G08B25/14	Central alert receiver or notification device	9

.

4.7. Keyword Frequency Analysis

We conducted a frequency analysis to examine the technologies for solitary death prevention in detail. The top 50 technologies are listed in

Table 11. Keyword frequency analysis of digital technology patents for solitary death prevention.

Rank	Word	Frequency (Number)	Rank	Word	Frequency (Number)
1	Communication	1000	26	Internet	55
2	Data	944	27	Remote control	51
3	Sensing	808	28	Robot	49
4	Terminal	768	29	Control	47
5	Monitoring	621	30	Speaker	46
6	Bio-	494	31	AR	46
7	System	395	32	Sensor	43
8	Image	257	33	LED	42
9	AI	246	34	CCTV	41
10	Voice	238	35	Sound	39
11	Pattern	231	36	Lamp	38
12	Wireless	161	37	Radio wave	36
13	IoT	156	38	Chatbot	32
14	Lighting	148	39	Algorithm	31
15	Remote	136	40	Infrared light	28
16	Application	133		Microphone	28
17	Network	100	42	Vibration	27
18	Shooting	94	43	Noncontact	24
19	Wearable	92	44	Cloud	20
20	Gateway	89	45	Deep learning	17
21	Radar	84		UWB	17
22	Radio frequency	79	47	GPS	16
23	Energy	76	48	Drone	12
24	Platform	63	49	Metaverse	11
	VR	63		AMI	11

. Communication, data, sensing, terminal, monitoring, bio, system, image, and AI were the top technologies.

4.8. Centrality Analysis

Degree centrality, which represents the extent to which a node is connected to other nodes in the network, is a measure of a node’s influence on the network. As shown in

Table 12. Centrality analysis of digital technology patents for solitary death prevention.

Word	Degree Centrality	Closeness Centrality	Betweenness Centrality	Eigenvector Centrality
Communication	0.9795	0.98	0.0536	0.2091
Data	0.9387	0.9423	0.0394	0.2064
Sensing	0.8979	0.9074	0.0359	0.1996
Terminal	0.9387	0.9423	0.0387	0.2065
Monitoring	0.9387	0.9423	0.0442	0.2048
Bio-	0.7142	0.7777	0.0138	0.1713
System	0.9999	1	0.0624	0.2102
Image	0.8367	0.8596	0.0262	0.1912
AI	0.7346	0.7903	0.0120	0.1789
Voice	0.7755	0.8166	0.0145	0.1858
Pattern	0.6326	0.7313	0.0070	0.1616
Wireless	0.8163	0.8448	0.0172	0.1934
IoT	0.5714	0.7	0.0035	0.1500
Lighting	0.6326	0.7313	0.0059	0.1632
Remote	0.6938	0.7656	0.0111	0.1728
Application	0.5714	0.7	0.0034	0.1502
Network	0.7346	0.7903	0.0142	0.1773
Shooting	0.6326	0.7313	0.0118	0.1535
Wearable	0.4285	0.6363	0.0028	0.1154
Gateway	0.5918	0.7101	0.0041	0.1544
Radar	0.6530	0.7424	0.0079	0.1619
Radio frequency	0.6122	0.7205	0.0078	0.1534
Energy	0.4693	0.6533	0.0018	0.1272
Platform	0.5510	0.6901	0.0027	0.1464
VR	0.2448	0.5697	0.0004	0.0670
Internet	0.6326	0.7313	0.0048	0.1635
Remote control	0.2448	0.5697	0.0001	0.0707
Robot	0.3469	0.6049	0.0012	0.0933
Control	0.6122	0.7205	0.0075	0.1539
Speaker	0.6734	0.7538	0.0108	0.1644
AR	0.2448	0.5697	0.0002	0.0654
Sensor	0.1836	0.5505	0	0.0577
LED	0.3469	0.6049	0.0001	0.1020
CCTV	0.4693	0.6533	0.0034	0.1215
Sound	0.4693	0.6533	0.0023	0.1232
Lamp	0.1224	0.5326	0	0.0359
Radio wave	0.5306	0.6805	0.0069	0.1295
Chatbot	0.3265	0.5975	0.0004	0.0940
Algorithm	0.3673	0.6125	0.0003	0.1040
Infrared light	0.3469	0.6049	0.0016	0.0959
Microphone	0.4693	0.6533	0.0017	0.1262
Vibration	0.3469	0.6049	0.0003	0.0950
Noncontact	0.3673	0.6125	0.0007	0.0985
Cloud	0.4285	0.6363	0.0009	0.1187
Deep learning	0.2653	0.5764	0.0000	0.0777
UWB	0.3469	0.6049	0.0004	0.0954
GPS	0.3469	0.6049	0.0015	0.0907
Drone	0.1020	0.5268	0	0.0318
AMI	0.1836	0.5505	0.0000	0.0500
Metaverse	0.1224	0.5326	0	0.0364

Note: The top 5 centralities are marked in blue.

, “system” had the highest degree centrality (0.9999); communication, data, terminal, and monitoring also demonstrated a high degree centrality. “System” also showed the highest level in the following: closeness centrality, which shows how quickly and effectively a node can access other nodes; betweenness centrality, which mediates interactions within a network; and eigenvector centrality, which is the degree of dependence on the influence of neighboring nodes. In addition to systems, communication, data, terminal, and monitoring ranked highly in terms of centrality. Figure 2 shows the network structure as a result of centrality analysis, meaning that the larger the node size and the more connected lines, the higher its importance.

4.9. CONCOR Analysis

As a result of exploring digital technologies for preventing lonely deaths through structural equivalence, eight key areas were identified, as shown in Figure 3 and

Table 13. CONCOR analysis of digital technology patents for solitary death prevention.

Cluster	Images per Cluster	Cluster Name	Keywords
1		Data and communication	Communication, Data, Sensing, Terminal, Cloud, Wireless, System, Internet, Remote, Application, Radio Wave, Wearable
2		Advanced monitoring	Monitoring, Bio, Radar, Noncontact, UWB, Metaverse, Vibration
3		Smart environmental control	Lamp, LED, Infrared light, Gateway, IoT, Sensor
4		Remote control	Remote control, Radio Frequency, Microphone, AMI, Lighting
5		Multimedia	Image, Speaker, Voice, Drone, Shooting, Sound, Robot, Control, GPS
6		AI	AI, Algorithm, Pattern, Platform, Energy, Chatbot
7		Extended reality (XR)	AR, VR
8		Learning-type surveillance	DeepLearning, CCTV

: data and communication, advanced monitoring, smart environmental control, remote control, multimedia, AI, extended reality (XR), and learning-type surveillance.

First, data and communication, which is a core technology for solitary death prevention, such as data collection through wearable devices and sensors and remote data transmission technology based on applications, the cloud, and wireless networks, were derived. This group was named data and communication. As the largest group, it is important as it enables a swift response to the risk of solitary death by analyzing information and patterns of vulnerable groups in real time and sharing bi-directional data.

Second, advanced monitoring technology is utilized to remotely monitor the health status of users to respond appropriately to dangerous situations. Monitoring, biosensing, noncontact sensing, and ultra-wideband (UWB) enable more accurate and sensitive identification of users’ status, while the metaverse improves the experience of users with limited mobility. Therefore, the group was named advanced monitoring.

Third, technologies were developed to maintain sufficient lighting and appropriate temperature through lamps, light-emitting diodes (LEDs), and infrared light sensors, and to actively respond to dangerous situations while monitoring users’ activities via gateways and IoT-based sensors. This is called smart environmental control, which automatically controls and optimizes the living environment. This technology is expected to improve the quality and safety of residential environments for users with limited mobility and the elderly.

Fourth, remote control technology was developed to easily control lighting, heating, and security systems in residential spaces through remote control, radio frequency, microphones, and advanced metering infrastructure. These technologies enable users with limited mobility to have independence and convenience in their daily lives.

Fifth, the multimedia group encompasses multimedia such as images, speakers, and voices. This contributes to making users’ lives safer and more comfortable, less isolated, and more socially connected. Drones, robots, and the Global Positioning System (GPS) are relatively new types of interface technologies expected to provide users with convenient and immediate responses through interactions.

Sixth, AI technology was applied to learn the energy patterns in residential areas and daily behavior patterns of users and automatically detect abnormal situations via AI algorithms and pattern analysis technology. Additionally, it is possible to alleviate loneliness and provide emotional support through AI-based platforms and chatbots.

Seventh, there was an active development of XR technologies that enable users to engage in real-life interactions and social connections for lower isolation. Such XR technology is expected to facilitate non-face-to-face communication, such as video chats or the sharing of residential spaces in virtual environments.

Eighth, CCTV systems using deep learning algorithms have been developed to learn user behavior patterns and send immediate notifications to prevent dangerous situations when abnormal behavior is detected.

5. Discussion

In this study, we performed an analysis to derive the rationale for the convergence of various technologies for solitary death prevention and the future technological impact factors of the isolated group with a high technological impact. Based on the current status and development directions of digital technologies for solitary death prevention, we derived the following technological development directions.

First, technologies for solitary death prevention were developed, mainly for risk monitoring and lower isolation. Risk monitoring technologies were designed to continuously monitor users’ daily life patterns to quickly and effectively respond to emergencies. In this field, advanced monitoring, AI, and scalable surveillance technologies were applied to preemptively detect and respond to risks that may arise in isolated user situations to minimize emergencies. Conversely, technologies to reduce isolation were developed to help users feel socially connected. To this end, technologies integrated with multimedia interfaces, XR, and AI technologies provide a wide range of diverse interaction services. These technologies have played an important role in alleviating isolation and enhancing social connections by enabling users to participate in social activities in virtual spaces and providing emotional support through conversations with AI chatbots.

Second, smart environmental control and remote control technologies were developed for users with limited mobility and those vulnerable to solitary death. These technologies were designed for residential applications and to automatically control lighting, heating, and security systems without requiring users to adjust them in person. They support more independent and safer living by making it easier to make the necessary environmental adjustments in everyday life in a facile manner. Smart home technologies have become an important means of reducing isolation by creating comfortable living environments.

Third, the convergence and integration of various technological fields have made solitary death prevention possible. Data-based intelligent monitoring, smart home environmental control, and user-friendly interface technologies are expected to play pivotal roles. The application of advanced technologies such as AI and XR will likely enable the development of cutting-edge prevention systems. To enhance the acceptance of these technologies among elderly users, it is essential to improve interfaces with user experience (UX) design in mind, provide educational programs, and continuously refine the technology based on user feedback. For technologies such as wearables, remote control systems, chatbots, and extended reality platforms, which rely heavily on user interaction, it is important to increase accessibility and ease of use by considering the diverse needs and abilities of the target age group. Additionally, personalized solutions should be offered to meet individual requirements.

Fourth, reliable and secure data processing technologies must be prioritized. Solitary death prevention technologies rely heavily on the use of personal information from elderly individuals living alone, single-person households, and patients, with data sharing and communication technologies playing a crucial role. Therefore, robust privacy protection and security measures must be meticulously established for systems, communication, and data processing technologies. Elderly individuals, in particular, tend to have lower awareness of security issues and may be less familiar with using digital devices or online services, making them more susceptible to cybercrime. This security vulnerability can heighten their anxiety and exacerbate feelings of isolation. Thus, when applying digital technologies to elderly populations, it is essential to provide accompanying services such as cybersecurity education, security monitoring, and support.

6. Conclusions

In this study, we collected textual data on patents and applied social network analysis techniques to analyze digital technologies for solitary death prevention. The findings of this study are as follows:

First, by examining trends in patents related to solitary death prevention through the IPC, we found a high rate of patent applications in data processing or coefficients, signaling, medicine and hygiene, and information and communication technologies. This suggests that integrated technologies for data processing and analysis, data transmission, and reception have been developed to prevent solitary deaths, with a particularly important role in the medical field.

Second, by examining the IPC main classification code of patents related to solitary death prevention, we found that G08B 21/04 (response to inactivity) was the most common classification. This implies that among the technologies for solitary death prevention, those that detect and respond to inactivity are the most important. They featured a convergence of alarm technologies and machine learning. Subsequently, G06Q 50/22 (social services or social welfare) was the second most common classification. This finding implies that technologies for solitary death prevention are closely related to social services and welfare. These technologies play an important role in providing individuals at risk of solitary death with social and emotional support in connection with social support systems. They featured the convergence of patient data management and diagnostic technologies.

Third, by examining the centrality based on patent content, we found that systems, communication, and data play a central role in the structural network of digital technologies to prevent solitary deaths.

Fourth, the CONCOR analysis showed the eight main areas for digital technologies for solitary death prevention: data and communication, advanced monitoring, smart environmental control, remote control, multimedia, AI, XR, and learning-type surveillance. These technologies have contributed to solitary death prevention by enabling the prediction of solitary death risks, real-time monitoring, remote control of residential environments, increased interactions and social connections, pattern analysis and prediction, lower isolation through VR and AR, and the detection of abnormal behavior.

Based on the findings of this study, the implications of digital technology in preventing solitary deaths are as follows.

This study is significant in that it explores in depth how various aspects of digital technologies for solitary death prevention have been developed through patent analysis. Digital technologies for the prevention of solitary deaths have two objectives: risk detection and reducing isolation. Accordingly, diverse technology fields such as data collection and analysis, vital sign monitoring, smart home control, multimedia interfaces, AI, and XR have converged. Furthermore, environmental control technologies have been developed to construct environments that provide users with psychological relief and lower risks and to enhance the convenience of users with limited mobility. This provides basic information for technology development, policies, and relevant studies to facilitate the development of technologies that fit the purpose and are more effective.

Despite the significance of this study, there are several limitations to acknowledge.

First, as this research relies on patent data, it has inherent limitations in demonstrating the actual effectiveness of the identified digital technologies in preventing solitary deaths. Future studies should address this limitation by conducting in-depth evaluations of the applicability of these technologies through field validation and case studies. Specifically, exploring and validating digital technologies that can be tailored to more specific aspects of solitary death prevention, such as health monitoring, elderly care, and management of single-person households, could lead to more practical and effective solutions.

Second, while this study identified the overall functions of digital technologies for preventing solitary deaths, it has a limitation in that it did not sufficiently identify technologies that can be applied to the segmented risk factors based on the different causes of solitary death. Therefore, future research should focus on further segmenting the risk factors according to the causes of solitary death and developing tailored solutions that address the specific needs and circumstances of each type to maximize the effectiveness of digital technologies.

Third, while patent analysis is useful in understanding current technological trends, it has limitations in predicting the actual performance and sustainability of technologies. Evaluating the sustainability of a technology is crucial to ensure its effectiveness and long-term success. Therefore, before committing resources and time to digital technologies for preventing solitary deaths, it is important to assess their efficacy and sustainability during the initial implementation phase through pilot testing. Based on these assessments, a cost–benefit analysis should be conducted to secure the long-term viability of the investment. Future research that validates the sustainability of digital technologies through the results of pilot tests and cost–benefit analyses will enable more efficient use of investment resources and time.