dagstuhl search as learning: summary breakout 1

9
Search as Learning 17092 Preben Hansen Claudia Hauff Claus-Peter Klas Kevyn Collins-Thompson

Upload: claudia-hauff

Post on 08-Apr-2017

132 views

Category:

Science


1 download

TRANSCRIPT

Page 1: Dagstuhl Search as Learning: summary breakout 1

Search as Learning17092

Preben HansenClaudia HauffClaus-Peter KlasKevyn Collins-Thompson

Page 2: Dagstuhl Search as Learning: summary breakout 1

Don’t forget the wiki

Page 3: Dagstuhl Search as Learning: summary breakout 1

Questions

1. What are the different types of learning (procedural, structured, incidental, formal, lateral …)? Should we treat them separately?

2. How can we help users that are not even aware of the terminology to jump-start their searches?

3. Is automatic scaffolding an option?

Group 1: Understanding search as a human learning process: when and how does learning occur in the search process?

Page 4: Dagstuhl Search as Learning: summary breakout 1

Questions

1. How can we address contextual gaps we cannot currently observe?

2. How does context interact with learning?

3. How can we tease apart multiple contexts that happen over time?

4. How can we deal with context latency?

Group 2: What is the relationship between the learning process and the context (educational, work-related, etc.) the learning & searches occur in?

$10,000,000

Page 5: Dagstuhl Search as Learning: summary breakout 1

A divide

▪ Have we reached a common ground?

▪ Do both sides understand/accept each other’s position?

▪ Have “we” (IR/IS) learnt what do wrong? Are we convinced? Can we learn how to do things right?

Group 3: How can constructs results from cognitive psychology and education be used to inform models of learning, knowledge acquisition, or mental representation during search interaction?

CognitionBloom

Page 6: Dagstuhl Search as Learning: summary breakout 1

Questions

▪ How generalizable are the often small-scale studies with specific search topics, specific tools and specific types of learning?

▪ Given the confounds, what are good proxies to measure learning (certain behavioural change over time)? Are they validated?

▪ Can we exploit approaches conducive to learning (e.g. retrieval practice) to measure learning?

▪ We often learn from not having achieved our goals? Measurable?

Group 4: How can learning performance and learning outcomes be measured during search? What search process features can act as indicators of learning?

Page 7: Dagstuhl Search as Learning: summary breakout 1

Questions

▪ How to move from a search box to an interactive conversational agent?

▪ How can we track what users do with the information after the search?

▪ When and how to target user learning during search?

▪ Can we encourage self-reflection on the search system interface level?

▪ What higher-level types of learning can by slower/human-in-the-loop processes?

Group 5: What type of functionalities and interventions on the search system interface level and the search algorithm level can foster learning?

Page 8: Dagstuhl Search as Learning: summary breakout 1

Today

We go with the flow! Changes are possible!

Page 9: Dagstuhl Search as Learning: summary breakout 1

Let’sstart!