Extraordinary improvements in medical science and other factors have resulted in greater life expectancy, thus rapidly increasing aging population across the world. The number of people over 60 years of age is estimated to increase to 1.4 and 2.1 billion by 2030 and 2050 respectively. Amongst many other problems experienced by older people, social isolation and loneliness are extensive.

  1. For instance, 20–34% of older people in 25 European countries and 25–29% in the USA reported being lonely.
  2. According to some study, 1 in every 3 people are lonely in some countries.
  3. A study in 2021 indicated a prevalence of loneliness of 25–32% in Latin America, 18% in India 3.8% in China.
  4. Other estimates of the prevalence of loneliness among older people were 29.6% in China, and 44% in India.

This has serious consequences for longevity, physical and mental health, and well-being. It is imperative we pay more attention to this issue and make the lives of those in need more comfortable and companionable.

UN and WHO suggest 'Reaching Out' to local services to connect with new people, communities, or professional help. In line with the same, our solution "In need, Indeed" is a single, scalable platform that will help senior citizens, who are lonely, left unattended, or need assistance, to reach out for companionship as per their needs and preferences while they would also be able to share their experience from various areas and valuable life lessons to the younger generation fostering an environment for human bonding, meaningful connection, and exchange of inter-generational ideas, lessons, and camaraderie.

What it does

“In need, Indeed” is a multi-channel platform that has the following functionalities:

Profile Registration – Both senior citizens and younger volunteers will be able to register their profiles, upload their identity proof, and indicate their preferences -interest areas (Reading books, walking, watching movies/sports, playing card games, etc), the language of communication, gender preference, time schedule

Profile Validation – System will check the uploaded identity proof to validate the legitimacy of the profile. In addition, system will automatically use the profile details to perform a sentiment analysis on the profile. This is to determine if the user volunteer is empathetic enough to engage in such activities.

Profile Matching – System will evaluate all the profiles of the volunteers and find a suitable match with the senior citizen based on a weighted scorecard comprising of the following preference criteria

  • Age range
  • Gender
  • Language
  • Interest area

In addition, system evaluates the profiles based on their proximity to the senior citizen and ranks them in the appropriate order.

Matching Initiation - Senior Citizens will be able to choose the volunteer with whom they want to pair up and initiate a pairing request. Once the volunteer accepts the request, system will pair the two of them.

How we built it

The solution uses Pega’s low code application development platform and its various features like Questionnaires, Integration capabilities (Google Cloud Vision OCR, Distance Calculation), NLP (Sentiment and Text Extraction), Business rule engine (Scorecard, Strategy and Prioritization Techniques), UI, and case management.

Please refer to Fig 1 for an overview of the various Solution components that have been used.

Challenges we ran into

Following are a few challenges we ran into during the development phase:

  • Creating new Text Extraction models using RUTA scripting language was new to us. We learned how to write RUTA script for detecting various entities e.g., PAN number, Date of birth etc, and how to set it up and test the model from Prediction Studio.
  • Integrating with Google vision API – we initially had trouble in finding out the right API to use, and what kind of billing mechanism to follow. We created a Google developer account and created a billing plan for a limited number of requests for using the API
  • Empathy Analysis - We wanted to use Pega Customer Empathy Advisor for analyzing the empathy of the volunteers in order to judge if they would be suitable for the role. However, we did not find any suitable reference for this and were not able to implement the same in the Community Edition. Alternatively, we used sentiment analysis using Text Extraction models for judging the positive and negative sentiments of the volunteers. In the future, we can extend this to integrate with other personality assessment services and also implement Pega Customer Empathy Advisor to make this a more robust solution.

    Accomplishments that we're proud of

    We are proud to have been able to blend into the solution various features of the Pega platform, including Prediction Studio to build Text Extraction models, scorecards, decision strategy, and prioritization techniques to create a scoring model. This has set the base for extending the solution to collect feedback and include the same as one of the parameters for making better recommendations and introduce various other capabilities as highlighted in the "What's next" section.

    What we learned

    1. We learned how to write RUTA script for creating entity-extraction models, and implement a decision strategy using scorecard and prioritization techniques for making recommendations.
    2. Sentiment Analysis - We have used Pega OOTB sentiment models for sentiment analysis of user through analyzing the answers to the provided questions, while developing this creating a text analyzer to fit this sentiment model and use this in the text analyzer through Lexicon. All those decision data parts were new to us..
    3. Integrating with proximity API - We have used the Google distance calculation API. After getting top 6 matching profiles, through this API we took out the profile that are nearest to the user.

    What's next for In need, Indeed

    This solution can be extended to support multiple geographies and languages and provide an enhanced recommendation based on user action and feedback received from both parties The inbuilt functionality of calculating credit points and engagement score for individual volunteers can be leveraged by different corporate entities to engage in CSR activities by means of providing eligible candidates with financial support/ scholarships etc. Following are some of the features that can be implemented to further enhance the solution:

    1. Online, offline Registration
    2. Connection triggered by Volunteers
    3. Feedback Collection for a better recommendation
    4. Feedback-based Credit Points for Volunteers
    5. Multiple Connection
    6. Helpline assistance
    7. Scheduling & Availability Assistant
    8. Professional Service Offerings
    9. Specialization by Geography & Language
    10. Corporate Tie up for CSR Activities
    11. Integration: Personality Profiling Services
    12. Integration: Background Verification

    Source: Advocacy Brief: Social Isolation and Loneliness, Fact Sheets: Ageing and health

  • Built With

    + 50 more
    Share this project: