Social Needs Marketplace

Project Team

Dionne Aleman
Chris Beck
Mark Chignel
Mariano Consens
Mark S. Fox
Wu Ga
Michael Gruninger
Chang Liu
Daniela Rosu
Yi Ru
Scott Sanner


Recent Publications

"The Semantics of Populations: A City Indicator Perspective"

"Increasing the Effectiveness of the Non-Profit Sector Through Virtualization: A Case Study of Furniture Banks"

"Knowdge-Based Provisioning of Goods and Services: Towards a Virtual Marketplace"

The Problem
Our project focuses on developing a more efficient and effective way to provide goods and services to the most vulnerable individuals within our society.  Of greatest concern are those citizens who live below the poverty line, the infirm and seniors -- those who often lack the funds, social support and/or mobility to access the social services they need.
Develop an online marketplace called the Social Needs Marketplace (SNM) in order to:
  • Provide a more efficient means of redistributing new and used goods and services by enabling the “demand” side to make known their needs, and the “supply” side to make known what they have available.
  • Address the logistical problem of transporting goods from the supplier to the consumer within an acceptable window of time. The Marketplace will provide a volunteer/agency supported by a logistical system that uses an Uber-like volunteer network.
  • Enable more efficient use of NGOs by virtually combining and focusing their resources to achieve better outcomes.
There are three major objectives that must be addressed in order to develop a Social Needs Marketplace:
1. Establish user profiles.  Consider a supply side user such as a grandmother.  We would like to learn the goods and services she can make available to the demand side. At the outset, individual user profiles are built by translating self-declarations (i.e., statements in English made by supply-side users that reveal what they are willing to supply) into a semantic representation.  Over time, we will estabish categories of profiles and identify the associated goods and services that meet their needs.  For example, new immigrants or new mothers form categories of users on the demand side, and users in the same category often seek similar goods and services.
On the supply side, empty-nest couples often no longer need their children’s old furniture.  There are also categories of users who appear on both the supply and demand side, such as parents who need to obtain sports equipment of different sizes as their children grow. When the demand-side user seeks a good, we can use the user profiles to actively seek out supply-side users who have not self-declared the good as available, but have a high probability of possessing it.  In other words, we do not have to rely solely on self-declarations for making potential matches.
2. Represent knowledge of users, goods, services, logistics, preferences, etc.  Any description of goods and services faces several daunting challenges. First, the terminology that people use for these descriptions is rife with ambiguity. Second, it is often unclear whether or not the terminology used is adequate to provide complete descriptions of a given range of goods and services. A third
challenge is that many of the terms that people use to describe goods and services are ad hoc and arbitrary. We need a knowledge representation into which natural language self-declarations can be mapped.
3. Overcome the challenge of physically transporting the donated goods from the pickup location to the delivery location.  Creating an online marketplace that will provide a more efficient means of redistributing new and used goods and services presents several challenges:
  • How do we use the online marketplace to match donated transportation services with requested pickup-and-delivery tasks?
  • How do we address the challenges of (1) geography and time (the driver will need to physically meet with both the supplier and consumer at convenient times without going far out of his/her way), (2) incentives and trust (how we establish a system where suppliers and consumers will be comfortable and safe in interacting with volunteer drivers), and (3) demand planning, prediction, and market clearance (how do we set up a system that will ensure that the required pickup-and-deliveries can be performed)?
Our approach to addressing these objectives applies and extends multiple areas of artificial intelligence. These areas of research include ontologies, machine learning and constraint-directed reasoning, operations research, human-computer interaction and database systems.
The project is divided into 9 tasks:
  • Task 1: Ontologies/Standards for representing goods, services, demand and supply
  • Task 2: Lexical Ontology Mapping
  • Task 3: User Modeling and Interaction
  • Task 4: Profile Learning for Enhanced Search
  • Task 5: Privacy and Fraud Detection
  • Task 6: Matching suppliers to demand individuals
  • Task 7: Pickup-and-Delivery Allocation
  • Task 8: Pickup-and-Delivery Prediction and Incentivization
  • Task 9: Ensuring Trust
Expected Outcomes
A major benefit of this project is the impact it will have on the quality of life of the vulnerable. The flexible and nimble provisioning of basic needs reduces the stress of many segments of our population. Reducing stress provides a better quality of life.  A better quality of life results in better work place performance and productivity.
Project Participants
  • Centre for Social Services Engineering
  • Findhelp Information Services (211 Toronto)
  • City of Toronto
  • Tata Consulting Services
  • IBM Canada