����ֱ��

We are developing AI technologies that are used in the context of STEM learning and are dedicated to nurturing effective, engaging and equitable classroom environments. Read more about our AI Partners and Tools below. For more information on how to incorporate our AI Partners into the classroom, please contact our team at Info.AI-Institute@����ֱ��.edu

Community Builder

The Community Builder (CoBi) AI partner helps groups of learners improve their collaboration skills by focusing on how they interact and work together. CoBi works by setting so-called "community agreements" — guidelines on how a learning community wants to collaborate, which include being respectful, equitable, committed to their learning community, and pushing thinking forward. Co-negotiated examples of community agreements are entered into the CoBi system to serve as a reminder of the community’s agreed-upon collaboration behaviors and as a reference for CoBi’s collaboration support.

CoBi uses microphones and language understanding to detect expressions of community agreements while learners engage in collaborative activities. These examples, or ‘noticings’, are then turned into various types of visual feedback that a learning community can use to think expansively about collaboration as a foundational skill and how it manifests in their co-located collaborations.

Jigsaw Interactive Agent (JIA)

The Jigsaw Interactive Agent (JIA) supports a specific type of group activity called a jigsaw, which is a cooperative learning strategy that asks groups of students to become “experts” on different aspects of a topic and then come together to share their knowledge with the group. Jigsaw activities encourage students to listen carefully to one another, discuss ideas, and build on those ideas to push the group’s thinking forward.

JIA offers two types of real time, interactive support: 1) tips to improve how a group is working together, and 2) hints for when a group is missing important information to complete an activity.

JIA is designed to work with minimal setup and little data, making it ideal for classrooms that are just starting to incorporate AI technology to support collaborations. The system uses a simple web platform that connects the jigsaw activities from the curriculum with a chatbot, which helps students talk through ideas and plan projects together.

The iSAT Recording Platform & Real Time Recorder

The iSAT Recording Platform is a web-based application that records video and audio and posts it to a data store and the iSAT Speech Processing pipeline. The web application is optimized for use on Chromebooks and produces a low bandwidth a/v format that is highly compressed, providing a low barrier for use by students and researchers in classroom and lab settings. The platform deploys to and operates from the Amazon Web Services (AWS) cloud.

Technologies used:  (S3 data store, Lambda, EC2, CloudFormations), ,  , , , , .

Requirements for use: AWS cloud account, MongoDB Atlas account, GitHub, knowledge of operating systems in AWS and the technologies used.

Please contact John Weatherley for more information and/or to schedule a consultation.

The Real-time Recorder is a stand-alone application that includes an audio recorder and a local instance of iSAT’s Speech Processing Pipeline, which can be installed on GPU or CPU laptop or desktop computers. Audio is captured and processed through the speech pipeline to provide a low latency, near real-time analysis of speech. The resulting analysis data can be saved to a database or output to a spreadsheet.

Technologies used: , python, . Optional:  (DynamoDB).

Requirements for use: Intel-based GPU-enabled laptop or desktop computer is recommended for best performance, but can run on Intel-based CPU computers as well. GitHub. Optional: AWS cloud account, knowledge of operating systems in AWS and the technologies used.

Please contact our team at Info.AI-Institute@����ֱ��.edu for more information and/or to schedule a consultation. 

MakeCode Activity Logging Platform

The MakeCode Activity Logging Platform is an expanded version of the blocks / JavaScript code editor for the micro:bit powered by  that has been extended to log the coding activity of users as they work. Each code edit is logged to a database in the MakeCode JavaScript format, which can be analyzed or input back into the editor to render the block code. The platform has two parts: The extended web based micro:bit block code editor and a back end data logging API server that logs the coding activity to a database.

Technologies used:  blocks/ JavaScript code editor,  (DynamoDB, ECS, EC2), python.

Requirements for use: AWS cloud account, GitHub, knowledge of operating systems in AWS and the technologies used.

Please contact our team at Info.AI-Institute@����ֱ��.edu for more information and/or to schedule a consultation.