Stitch Co., Ltd. has released "Coco-senior," a website for introducing senior housing facilities. The feature is the use of AI systems to match residents with facilities where they will be happiest. The AI was developed by AVILEN, Inc. Using data from over 1,000 actual tenants, AVILEN's recommendation engine "FALCON" learns and suggests the most suitable facilities for prospective tenants. CareTalk, Inc., which specializes in the nursing care business, has also joined the project as a business partner, and the service incorporates a wealth of knowledge from the nursing care field. The matching AI will continue to accumulate further data and learn to achieve higher accuracy.

Overview

• According to the results of the survey, the most important factor in selecting a facility for parents was "the atmosphere of the facility" at 53.4%.
• We analyzed data from over 1,000 actual residents to create an AI based on the atmosphere of the facility that best fits the residents. The AI analyzed data from over 1,000 actual residents and found that the closer the resident's "normal home attire" was to the "daytime attire of other residents at the facility," the happier the resident would be after moving in.
• The "correct answer rate" for "facilities that are assumed to have a high propensity to make residents happy" presented by the AI recommendation engine is 75%.
• Provides a "happiness support system" that eliminates post-occupancy mismatches by combining the introduction of facilities with a high likelihood of happiness using AI prior to moving in with monitoring of post-occupancy happiness.

The results of the survey showed that the most important factor in selecting a facility for parents was not cost or location, but the "atmosphere of the facility," selected by 53.4% of respondents (Figure 1). The importance of atmosphere in senior housing facilities has also been shown in terms of living space and interpersonal relationships (*1), but the current facility referral sites do not emphasize the importance of atmosphere.



In addition, although residents visit facilities to determine the atmosphere, it is difficult to find a facility that truly matches their life after moving in after only one visit. Coco-senior sees this as a social issue, and has built an AI that analyzes the atmosphere of facilities that best matches the residents, based on data from over 1,000 cases of people who have actually chosen a facility. This analysis yielded interesting data on the elements (food, clothing, shelter, and people) that make up the atmosphere. For example, the closer "the way residents usually dress at home" and "the way other residents dress during the day at the facility" are, the higher the level of happiness after moving in.



By analyzing the personality of the residents and facility data with this constructed AI, we will introduce facilities that have a high possibility of making residents happy. Furthermore, we will conduct a questionnaire to measure the sense of happiness at the time of the decision to move in, and 2 months and 6 months after moving in (*2). (*2) In addition to the items measuring happiness, factors that contribute to happiness are extracted from various research results and incorporated into the questionnaire, such as "whether you feel the staff at the facility smile," "whether there is someone to listen to your worries and complaints," and "whether you participate in activities. Based on the results, we analyze the factors that are increasing or decreasing the level of happiness, compile them into a report, and send it to the facility and family members who have moved in. If the level of happiness is decreasing, we will propose improvement measures.

In this way, we support happy living by introducing you to facilities that are likely to make you happy before you move in, and monitoring your level of happiness after you move in. Coco-senior offers this combination of selecting facilities that lead to happiness before moving in and monitoring the level of happiness after moving in as a "Happiness Support System" to all of our clients. This will reduce the possibility of mismatches with facilities after moving in.

About AI Algorithm

AI presents facilities that are likely to make residents happy based on their personal information. The process consists of three steps: 1. data input, 2. analysis by algorithm, and 3. presentation of facilities.

1. Data will be input in the form of answers to various questions about prospective tenants, such as "number of rooms you live in," "hobbies," "how you dress when you are at home," "type of work you have done for a long time," and so on.
2. The input data is analyzed by a proprietary algorithm using the AI recommendation engine "AVILEN FALCON. (AVILEN FALCON is trained in advance using a large amount of data on facilities and past tenants.)
3. As a result of the analysis, we present facilities that are assumed to have a high tendency to make the residents you are considering happy. (At present, the "correct answer rate" for "facilities that are assumed to have a high propensity to make residents happy" is 75% *3; further data accumulation and learning will make the system more efficient in the future.)

Learn more about AVILEN's AI development solutions
https://avilen.co.jp/service_category/solution/