Home
Explore Uedu
Student Console
Register as Member/Login
(2) In future presentations of the research findings, in addition to the course project website and public presentations, your real name and personal information will not appear in this research report. If you are interested in the research results, we can provide you with an executive summary after the study is completed.
問卷中心
Teacher Console
Course Setup
Support & Messages
Uptime Data

UeduGPTs

--

Jupyters

0

Local AI

--

CISOSE26 本地 AI UG26
政治大學 AQI 20 29°C

AI Reply Desktop Notifications

Show a desktop notification when the AI TA finishes replying

Chat Message Notifications

Notify me when classmates post messages in the forum

Sound notification

Play an alert sound whenever there is a new notification

An Educational Practice · Designed for Reproducibility

Educational Omics

教育組學 — An educational practice that operationalizes Multimodal Learning Analytics (MMLA)

EO 是 MMLA paradigm 在大學通識教育場域的具體 educational practice。在持續實踐中浮現 Cognomics、Linguomics、Physioneuromics、Sociomics、Environomics、Ethicomics 六個可觀測維度,並對 MMLA paradigm 反向回饋。Uedu 平台為 EO 的 first instantiation, 目前部署於 48 機構、累積 280,000+ learner-AI interactions。

MMLA → EO → Uedu

從 paradigm 到 practice 到 first instantiation,EO 的三層學術定位

PARADIGM

MMLA

Multimodal Learning Analytics

The broader learning analytics ecosystem since the 2010s. Learning behavior should be observed through multimodal data—voice, physiology, eye-tracking, interaction, text.

Venues: LAK · EDM · AIED · JLA · IJ-CSCL
Researchers: Worsley · Blikstein · Di Mitri · Cukurova · Martinez-Maldonado
PRACTICE

EO

Educational Omics

An educational practice that operationalizes MMLA in higher education general education contexts. Six omics dimensions emerged from sustained practice, designed for reproducibility.

Cognomics · Linguomics · Physioneuromics
· Sociomics · Environomics · Ethicomics
FIRST INSTANTIATION

Uedu

Agents / Platform

ClassroomGPT · AIDA · Garmin BBI streaming · Environomics CWA + MoEnv integration · 學習特質探索 · Educational Omics Lab.

EO can be instantiated on other platforms; Uedu is the first. Replication by other institutions is anticipated as future work.
PARADIGM  →  operationalized in practice  →  PRACTICE  →  first instantiation  →  PLATFORM

Important positioning:EO 不取代、不凌駕 MMLA。EO 是 MMLA paradigm 內部的 educational practice。透過持續實踐,EO 對 MMLA paradigm 反向回饋——明確命名 Ethicomics 與 Environomics 兩個 underdeveloped 維度、引入 biology multi-omics integration metaphor 作為跨模態整合方法論。

Operationalizing MMLA — 5-Stage Practice Pipeline

EO 在大學通識教育場域的具體實踐流程,從多模態感測到個人化學習

1
多模態
感測
收集學習者的 HRV、EEG、fNIRS、語音、互動紀錄及 IoT 環境感測數據。
2
Educational
Omics 分析
透過認知、語言、生理、社交、環境與倫理六大維度框架分析學習歷程。
3
AI 驅動
整合
Uedu 平台整合多模態數據,優學伴 AIDA 提供情境感知的學習支持。
4
適性化
介入
透過 PALM 生理感知語言模型,即時提供適性化鷹架輔導策略。
5
個人化
學習
以學習者為中心的大規模個人化學習,讓每位學生按自己的步調成長。
Educational Omics 六大組學維度
Cognomics
Bloom 認知層次、MBTI、Holland RIASEC
Linguomics
語音錄製、雙語字幕
Physioneuromics
HRV、EEG、fNIRS、穿戴裝置
Sociomics
討論區、協作編輯、校園留言板
Environomics
IoT 感測器、IP 攝影機、STDB
Ethicomics
IRB 倫理審查、資料治理、知情同意

六大組學維度

每個維度都揭示學習過程的不同面向;標 者為 EO 對 MMLA paradigm 的反向貢獻

Cognomics

認知組學

追蹤思考歷程與認知發展,包括 AI 對話軌跡、問題解決策略、Bloom's Taxonomy 層次分析等。

Linguomics

語言組學

分析語言表達特徵,包括語言複雜度、語意結構、語音特徵、書寫模式等。

★ EXTENDED

Physioneuromics

生理神經組學

監測生理狀態與神經活動,包括 HRV、睡眠品質、壓力指數、專注度。對 MMLA 補強:MMLA 多侷限於 EEG / GSR 短時實驗,EO 擴展至 long-window wearable 數據。

Sociomics

社交組學

分析社會互動模式,包括討論區參與、協作學習行為、同儕互評等社交學習數據。

★ EO CONTRIBUTION

Environomics

環境組學

感測學習環境條件,包括光照、溫濕度、噪音、CO2、AQI、PM2.5 等。對 MMLA 補強:MMLA 通常把環境視為 confound,EO 將其升格為 first-class observation 維度。

★ EO CONTRIBUTION

Ethicomics

倫理組學

觀測學習者在學習活動中的倫理面向行為訊號——consent 狀態與變動、AI 使用揭露完整度、override AI 建議的模式、對 biased 輸出的 challenge 行為、引用與歸屬實踐。對 MMLA 補強:MMLA 通常將 ethics 視為外部 IRB 議題,EO 將學習者的倫理行為內化為 first-class observation 維度。Uedu 平台層級合規(含 NTU-REC 202507EM058 核准)為 evidence base 之一,但與 learner-level ethics signal 屬不同分析單位。

查看 IRB 框架揭露

Evidence Base

EO 作為 sustained empirical practice 的累積證據

48
institutions

大學部署 Uedu 平台
(platform reach,not yet EO replication)

280k+
interactions

Learner-AI interaction traces 作為 multimodal evidence

6
omics dims

Practice-emergent stratification(非 a priori 設計)

NCU
first instantiation

Educational Omics Lab @ 國立中央大學

Note on reproducibility: Uedu/NCU is the first explicit EO instantiation. The 48-institution deployment demonstrates platform reach; broader replication of the EO practice across institutions is anticipated as future work. EO is designed for reproducibility through open API, MCP servers, and Educational Omics Lab consulting.

EO 對 MMLA 的補強

透過 sustained practice,EO 對 MMLA paradigm 的具體反向回饋

MMLA 現狀 EO 補強(practice-emergent)
「多模態」無固定 taxonomy,每個 paper 自己選模態組合 六個 omics 層作為 reference stratification(實踐回溯浮現,非 a priori)
Ethics 通常被視為「外部 IRB 議題」,不在分析框架內 Ethicomics 把 ethics 訊號 內化為觀測維度
Environment 變項通常被當作 confound 處理掉 Environomics 把環境視為 first-class observation(光、溫、噪音、CO2、AQI、PM2.5)
Physiological data 多侷限於 EEG / GSR / 眼動短時實驗 Physioneuromics 涵蓋 long-window wearable(HRV、sleep、stress、daily heart rate)
跨模態整合多為 ad-hoc fusion 引入 biology multi-omics integration metaphor 作為跨模態整合方法論
缺乏 reproducible 的 large-scale practice case Uedu 平台部署於 48 機構、280k+ interactions,作為 reachability evidence(broader EO replication 為 future work)
學習評量主流仍依賴 summative assessment(exam, paper),在 LLM era 失去 differential validity——AI-generated 產出與 learner-authored 產出難以區辨 Cognomics dialogue trace 提供 formative assessment 的 process-level evidence base(呼應 Black & Wiliam 1998, Wiliam 2011)——可被自動標記至 Bloom's revised taxonomy 六層,使 formative assessment 在 scale 上可操作;補強 summative assessment 在 GenAI 時代的盲點,兩者並存

所有補強皆 within MMLA paradigm,不在外面。EO 是 MMLA paradigm 內的 educational practice,而非取代或凌駕 MMLA。Reverse contribution 的目的是在實踐中讓 MMLA paradigm 更完整。

Uedu Platform as First Instantiation

EO 在 Uedu 上的具體子系統。EO 可被其他平台 instantiate,Uedu 為 first case

ClassroomGPT

AI 蘇格拉底對話,收集 Cognomics / Linguomics 數據

了解更多
Uedu Fit

Garmin 穿戴裝置整合,收集 Physioneuromics 長窗時序數據

了解更多
Environment Module

CWA + MoEnv 整合,收集 Environomics first-class data

了解更多
Uedu Labs

研究者資料介面,存取 Trusted Educational Data Lake

了解更多

Engage with Educational Omics

EO is an open, designed-for-reproducibility educational practice. Researchers, institutions, and MMLA-aligned collaborators are invited to engage.

AI Literacy 6D Framework Research Ethics Access Data Lake