Thesis - Part II

4 Jan 2017

So, what’s my thesis about? Supporting teachers to manage semantic diversity in the classroom in inquiry-based learning scenarios. That’s quite a title (temporary, by the way). The research group I’m working with is focused in the big field of Learning Analytics (LA), so let’s first talk about (Learning) Analytics.


I’m sure you’ve heard about the term Analytics before (e.g. Google Analytics, Adobe Analytics, software analytics); basically, it’s a tool organizations can use to interpret their big amount of data and get some benefits from it: marketing optimization, processes enhancement, patterns in the usage of some especific piece of software, etc. The term became trendy in November 2005 when Google launched Google Analytics (GA) and took over the Analytics market in less than a month offering a freemium service that tracked and reported website traffic1.

Learning Analytics (LA)

So, recollect, analyze and report. That’s the basic cycle the Analytics processes follow whatever the field it is applied to. In relation to knowledge, teaching and learning, institutions can make use of the data learners throw off in the process of accessing learning materials, interacting with educators and peers, and creating new content 2. Recently MOOCs have emerged as a popular mode of learning and reveal the need for specific tools for analyzing big amount of data.

The research group I’m working with is focused in the LA field and among other things they’ve developed a series of tools in the context of the Go-Lab Project which opens up online science laboratories for large-scale use in school education.

The overall aim of the Go-Lab Project is to encourage young people aged from 10 to 18 to engage in science topics, acquire scientific inquiry skills, and experience the culture of doing science by undertaking active guided experimentation. To achieve this aim, the Go-Lab project creates the Go-Lab Portal allowing science teachers finding online labs and inquiry learning applications appropriate for their class, combining these in Inquiry Learning Spaces (ILSs) supporting particular lesson scenarios, and sharing the ILSs with their students. Using the ILSs, the students receive the opportunity to perform personalized scientific experiments with online labs in a structured learning environment 3.

One of the apps the research group created is the Concept Cloud (CC). The CC offers a structured visual representation of semantic concepts based on a group-oriented model student knowledge. This model is extracted from the material the learners have produced during the inquiry phases (Wiki articles, concept map tools and hypothesis tools). The main goal of the app is to support learners in reflecting their own knowledge (the CC is displayed at the end of the inquiry process and learners can go back to improve their material if they think the CC doesn’t represent what they intended), as well as teachers in supervising their students during the process of learning.

Semantic tools

One of the core foundations of the CC is that it extracts semantic information from the material the learners have produced. Each artefact consists of a set of knowledge items that refers to specific concepts in the domain of the learning space. The extraction of these concepts is performed via semantic tools such as AlchemyAPI and DBPedia Spotlight; in this way at the end of the day we transform a Wiki article produced by the learner into a set of semantic key concepts contained in that article.

The CC app is fine as a visualisation tool, but it can be improved by analyzing its semantic features. One thing teachers would like to know is “are the learners using consistently the key concepts that are relevant for the topic under study?”, and if not, “which students deviate the most from the rest?”. Furthermore, the current layout of the CC doesn’t provide any semantic information about its items: they are placed randomly on the screen.

The topic of my thesis aims to support teachers in managing the semantic diversity of the classroom using the CC data model as input:

  • Teachers can know in which grade the CC produced by the learners is consistent. The more semantically diverse the CC is, the less consistent the concepts produced by the learners are.

  • With this semantic information, it’s possible to provide a new layout for the CC in which semantic related terms appear closer in the cloud of words.

For the next entry: semantic diversity in depth.

  1. Nowadays it does a lot more (review online campaigns, identify poorly
    performing pages, track sales activity and performance…).