Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 4 Current »

✏ Current To-Dos

  • Make a Schedule for Pre-Course work
 ℹ️ Pre-Course Schedule

  • Look at the new Bot in depth
  • Analyze student and TA feedback
  • Revisit rubric
  • Finalize assessment methodology - presentations?
  • Look into GitHub classroom options
  • Add more Supple/Marbel documentation
 ℹ️ What is Missing in Documentation?

  • Take extras out of the bot on branch cap. basic
  • Push new bot to school solution
  • Look into Marbel Plugin and Installation tricks.
 ℹ️ VS Code Extenstion instead of MARBEL Eclipse Extenstion??

who to harass for this?

  • Assignment restructure by function
  • Check assignment with new code.

✏ To-Decides

  • Assign Teams? - Idea mention in MAS offer spread sheet those not in will be assigned
  • Decide on Assignment Content (What is given)
 ℹ️ Suggestion

Give base bot instructions, recommend recipe, 1 filter example (country), 1 tag or ingredient type etc. have them decide and do rest. We have overview could give them an outline of how to do it but no instructions. Filters to be approved by TAs in a session

  • Finalize Bot Goals Overview
 ℹ️ Bot Goals

What do we want our overall bot to do?

  • greet

  • self identification

  • start

  • appreciation

  • recommend a recipe randomly

  • describe capabilities

  • specify goal

  • feature inquiry

  • say what is not in capabilities “can you do this? no”

  • farewell

  • conversation repair

    • unknown phrase, paraphrase request

    • context mismatch not appropriaet

  • filter recipes based on: country, name, ingredients, time, number of ingredients, number of steps, diet, easy/not easy, ingredient type, servings, quick, tag

  • filter recipe by exclusion: ingredient, dietary restriction, cuisine

  • provide filtered recipes with n recipes

  • remove features based on no recipes

  • show filtered recipes on request

  • check/confirm recipe

  • welfare check

NOT IN BOT:

  • performance rating

  • instruction

  • recipe rating

  • small talk with ChatGPT???

  • Mode of grading - presentation?
 ℹ️ Grading Suggestion Overview

4 Weeks. Turn In: Report at End, 2 Labs per week one lecture

Each week: Lecture, Lab, Half Lab/ Half CheckIn Presentation except first week no check in pres, last week no lecture and final presentation

Idea 12 teams per room. 3/4 TAs

Labs: Help from present TAs in the room, Goal: 1 TA per 3/4 groups in the room.

Check-In Presentation: 10 minutes total 7 minute student presentation 2 minutes questions from TA, 1 min TA feedback (2 hours min per 12 groups)

Graded on: Time of Presentation, Presentation Skills, Pres Content, Work Distribution, Answers to Questions

Pres content overview: Given ahead of time based on week content but essentially

  • Demo of bot

  • What was done divided by person.

  • Why it was done that way/ benefit

  • Challenges/Difficulties Encountered (Helps TA Documentation too)

  • Future Goals/Ideas

Questions from TAs most given ahead of time based on week content some group specific

Questions could be about:

  • What does _ do or is used for, why?

  • Explain a concept

  • Ask for clarification or expansion on a topic

  • Give Feedback on their topic - course correction on project

If someone is not present: submit individual! written report on topics of presentation like logbook entry can makeup 1 presentation with logbook otherwise 0 for that person for that week

Maybe extra bonus points based on oh they are doing awesome. All 3/4 TAs grade each group and ask questions, grading averaged over TAs

Longer final presentation?

End Report:

same as last year but experiment on

comparing their filters to one another by own metric for an experiment via use - similar lab experiment session ex. user satisfaction/difficulty metric, time to recipe selection, accuracy

Final Grade: 25 percent final bot product, 25 percent report, 25 percent final presentation, 25 percent check in presentations

  • Integrate Chatgpt?
  • add back instructions?
  • Python version of SIC or no?
  • How to split students
 ℹ️Split Suggestions for Fairness

Groups not assigned to TA but room by number. TAs rotate rooms each week.

Note Ideas:

  • instead of just easy make classification functions easy, medium, hard

  • add exclusion by tag? “ I do not want pizza” for them to do

  • add capabilities intent dialogflow

  • make them add spicy themselves

  • let them filter by what they want give single example for how to filter by a few things like nr of ingredients

  • Idea: instead of quizzes in canvas in person quiz by team for bonus points!!! live in first lab also great introductory

✨ Next To-Dos

  • SetUp TA Training
  • Give TA Training
  • Make school version of new bot
  • redo assignments for new bot version
 ℹ️ More info

  • No labels

0 Comments

You are not logged in. Any changes you make will be marked as anonymous. You may want to Log In if you already have an account.