Burglary Victims' Support Bot

"Smashing find, son!"

Why tackle burglary?

Burglary affects 2 in every 100 households, and it sucks! It's a miserable and frightening experience for the victim, and a huge drain on police resources.


* the period from Oct 2016 to Sep 2017

Police simply don't have the resources to reach every burgled home, or to investigate every crime. Victims of burglary might wait a few hours for a response, or a few days - if it is possible to send someone at all.

We want to help put that right!

We are building a conversational tool for the victims of burglary. We want to give them control over their own investigation, and save hours of police time as we do.

If each use of our tool saves 1 hour of police time doing initial investigation, then in a year we could save up to 664,000 police hours. Officers can spend that time helping, protecting and reassuring people.

Recent meet-ups

We held our first project meet up on the 12th March, in London.

Keep an eye out for more!

Project progress

Phase 1

  • ✅ Early design sketches
  • ✅ Initial framework
  • ✅ Opening conversation (learn user details)
  • Prepare a full set of early design documents
  • Design supporting victim database service
  • Stub out main flows
    • Record new burglary
    • Review and edit previous burglaries
    • Fulfilments

Phase 2

  • Implement supporting database
  • Implement main flows
    • Record new burglary
      • Enumerate losses (identify items, costs, serial #s, etc.)
      • Identify forensic opportunities
      • Examine insurance options
    • Review and edit previous burglaries
      • Overview all personal incidents
      • View details for a personal incident
      • Modify or correct a detail in a personal incident
    • Fulfilments
      • Generate personal victim statement
      • Examine home for prevention advice

Phase 3

  • Language Understanding for user interruptions
  • Conversational resilience testing
  • User acceptance testing

Phase 4

  • Trials with a police force
  • Sponsorship for live implementation

Contributing to the project

Join us and help build a conversational tool for people when they need it the most.

We'd also be grateful for your time if you've been the victim of a burglary yourself. Your experience is invaluable in helping us strike the right tone, and eliciting the most helpful things we can do for others in your predicament. Our bot must be courteous, informative and meet as many of the needs you had.

Designers

  • We'd love to work with someone passionate about this topic, and willing to help us create a friendly design aesthetic, and presence, for the bot.

Web creators

  • The bot needs a real, friendly presence on the web, too. Help us put together an accessible and helpful set of web pages that include a web control for the bot itself.

.NET coders

We're using the MS Bot Framework (v4) for a few reasons:

  • It's cross-compatible with many channels of communication.
  • We can integrate with external services.
  • We can easily build and test locally.

When the bot is ready, we'll be hosting on an Azure-like service (TBC). Until then, we're using the MS Bot Framework Emulator.

All the code is available at our github repository for the project:

Security awareness is a key part of every design and implementation decision. Bring your infosec tin-foil hat to every project involving personal data! We'll be challenging each other, our designs, and our implementations throughout the process; and we'll be seeking advice from security professionals as we work.

Conversation flows

This early sketch illustrates the flow of a conversation from start to finish.

This is commonly known as the happy path - where a user experiences everything as the designer intended without any interruptions.

Overview

The overview is derived from the user needs, details identified during work on the happy path, and a view of which tasks a user might want to complete during a session with the bot.

Credit for Hack the Police 3 poster: HappyToastCredit for illustrations: Lukasz Adam