1 of 12

MOSAFELY.ORG

PFI-RP: A Multi-Disciplinary Approach to Detecting Adolescent Online Risks, #1827700

Pamela Wisniewski (PI), Munmun De Choudhury (Co-PI), and Gianluca Stringhini (Senior Personnel)

Modus Operandi Safely (MOSafely)

Because we all want our youth to engage online more safely . . .

2 of 12

MOSAFELY TEAM

Pam

Munmun

Gianluca

Afsaneh

Matt

Shiza

Ashwaq

Xavier

Nurun

Faculty Team

Ph.D. Students

Svetlana

Andrea

Tech Transfer

Mozilla

Temi

3 of 12

MOSAFELY: STRATEGY TOWARDS COMMERCIALIZATION

  • Mission Statement: An open-source community that leverages evidence-based research, data, and Artificial Intelligence (AI) to help youth and young adults engage more safely online.
    • Will offer a paid customizable SaaS API for online platforms for detecting online risky content and/or risk behaviors, as well as indicators of mental health and well-being in the online interactions of youth and young adults.
  • The SaaS API will allow the open-source community to self-sustain and eventually incorporate as a non-profit organization

4 of 12

R&D OBJECTIVES

  • An open-source community (OSC) that contributes research, data, and untrained algorithms for adolescent online risk detection
    • To gain market visibility and broad participation, so that developers can train the algorithms with their own data sets, contribute code, and expertise.

  • A commercial Software as a Service Application Protocol Interface (SaaS API) based on our technology develop and integrates with code libraries from the OSC
    • Combines these algorithms modularly into an easy-to-use and accessible service for adolescent online risk detection.

5 of 12

POTENTIAL PRODUCTS AND SERVICES

Innovation: Adolescent Online Risk Detection

Software as a Service

Open-Source Community

Integrated Risk Assessment Tools

Customer Segments:

Safety Officers at Online Companies that have Teen Users

Developers and researchers who want to use/improve adolescent online risk detection

Social service providers and clinicians that work with teens

Beneficiaries: Parents and Teens

“AI Inside” Ingredient Product

for Product Integration

Web Dashboards, Mobile Apps, etc.

Domain expertise, Code Libraries, & Data

6 of 12

MOSAFELY.ORG CONTRIBUTION TYPES

Trained Algorithms/ SaaS API

    • Minimum Viable Product (MVP) a SaaS API that encapsulate trained and customer validated risk detection models. Proprietary Intellectual Property (Patents & Registered Copyrights)

Code

    • Code contributions of risk detection algorithms libraries. Untrained algorithms only. Shared through the �MIT License.

Data (Shared & Private)

    • Data Donations: Data donated by youth and stored privately by MOSafely (Restricted)
    • Training data sets shared by contributors shared through a Hippocratic Data Use Agreement

Forum

    • A community-based forum for researchers and developers to share knowledge relevant to the mission MOSafely.org

Research

    • Peer-reviewed research related to adolescent online safety and risk detection
    • Public access to evidence-based research

Public Access

Proprietary to MOSafely

Open-Source Community (Contributors Only)

7 of 12

MOZILLA & OPEN LEADERSHIP

8 of 12

MOSAFELY OPEN-SOURCE COMMUNITY

Launched Website

    • http://www.mosafely.org/

Created GitHub Organization

    • https://github.com/mosafely

Working through Legal Agreements

    • Contributors, Privacy, Licensing

Establishing Open-Source Community

    • Join Now!

9 of 12

MOSAFELY.ORG WEBSITE

10 of 12

SPRING 2021 STUDENTS: �MASTERS & UNDERGRADUATE

Presented by Pam

Christian

Sunil

Pallavi

Avery

Carlson

Komila

Mari

Benjamin

Abdul

James

Micah

Natalie

Omar

Zilong

11 of 12

SUPPORTING STUDENTS AND �PAST STUDENTS

Heidi

Karla

Zainab

Neeraj

Irina

Zach

Gabriel

Natalie

Amy

Tara

Kevin

Sahana

Hannah

And More!

Presented by Pam

12 of 12

THANK YOU! QUESTIONS?

  • Dr. Pamela Wisniewski (PI), Associate Professor, pamwis@ucf.edu
    • University of Central Florida, College of Engineering and Computer Science
    • Socio-Technical Interaction Research (STIR) Lab: http://www.stirlab.org

Join Now!