Introduction

Recent advances in social semantic web can offer the technical underpinning for taming digital traces; allowing semantically augmented user generated content available via semantic data browsers. However, for learning environments, new intelligent techniques are needed to extend semantic data browsers with features that facilitate informal learning, yet preserving the exploratory nature of social environments. We propose a novel approach for taming digital traces in social spaces for in-formal learning by combining major advancements in semantic web with 'semantic nudges' in a semantic data browser - called I-CAW.

An Intelligent Content Assembly Workbench (I-CAW) is a semantic data browser with intelligent features that facilitate informal learning. I-CAW utilises: (i) ontological underpinning and semantic augmentation and semantic query services to aggregate and organise digital traces, and (ii) nudges such as signposting and prompts to guide users through the browsing process. Following figure outlines the main component of I-CAW.

I-CAW processing pipeline

Digital Traces Collection

The availability of social web APIs has made it possible to consume digital traces from these sources and to build custom applications. I-CAW supports trainers by offering the options to browse the videos and comments from YouTube (linked within I-CAW) and personal stories collected from a blog-like story telling environment. Trainers can then select appropriate content for training material.

Ontology Underpinning

The Ontological Underpinning is crucial for aggregating digital traces on an activity, as a semantic model describes the key aspects of that activity in the form of an ontology. An Activity Model Ontology (AMOn) is developed in ImREAL by a multi-disciplinary team of computer scientists and social scientists. AMOn is described here .

Semantic Augmentation and Semantic Query services

Semantic augmentation uses ontologies to enrich unstructured or semi-structured data in relation to the context of an activity. Semantic query provides a mechanism for querying and browsing the semanti-cally augmented content in the repository. Both services are described here in detail.

Semantic Nudges

I-CAW proactively suggests areas of exploration to learners. We have developed a novel approach - nudges - based on Sunstein and Thaler’s choice architecture and a proposal for its adoption in learning. In a choice architecture, a choice architect is responsible for organizing the context in which people make decisions. A nudge is any aspect of the choice architecture that alters people’s behaviour in a predictable way without forbidding any options, and tries to influence choices in a way that will make choosers better off. Two types of nudge have been chosen: signposting and prompts. Semantic nudges are described here in detail.

User interaction Scenario

To illustrate how I-CAW could be used to make sense of real-life experiences from the digital traces, let us consider the following learning scenario. Jane is new to an organization and is commencing her first job in a human resources department. Her new role will involve interviewing applicants for a vacant position. After inquiring she was advised by some colleagues that interpersonal communication aspects such as non-verbal behavior can play a crucial role in the conduct of an interview and it is important that she is aware of them to help her successfully perform the interview. She wants to learn more about this.

She knows that there is a large volume of digital traces (videos and comments, blogs, stories etc) on the Web but it is time consuming to search this content and create meaningful associations between the content and draw inferences. Using I-CAW she searches the term “body language” and is offered several digital traces in the form of videos on YouTube (and comments) on job interviews, people’s real experiences represented in blogs, amongst other forms. This hybridization of heterogeneous content is made possible with the semantic augmentation service, which extracts key terms from these different textual contents and links them to ontological concepts, stores them in semantic repository and makes them available for querying.

In I-CAW Jane is able to browse through the recommended videos, comments and other digital traces. While reading about a particular blog, she learns that eye-contact can be important and that there may be several possible interpretations. She clicks on eye-contact and is particularly interested in the link pointing to “Body Language Meanings” (derived from the ontology by I-CAW), which leads to a list of manifestations of eye-contact (gaze, stare, and its interpretation as meaning anxious or nervous etc). From there, she clicks on “Nervousness” and arrives at another page with a collection of related YouTube videos, personal stories and comments. From the information she discovers other types of body language signals (other than eye-contact) that may indicate nervousness (derived from the relationships defined in AMOn ). During this process of inquiry, Jane learns from the diverse comments about the different interpretations of people from different cultures and the different perspectives (e.g. interviewer vs. job applicant, novice vs. experienced individuals).

In this scenario I-CAW has intelligently coalesced and made sense of different digital traces by exploiting the ontology and semantic augmentation. Users can develop insights into the phenomenon based on previous experience and insights of others as captured by the digital traces.

Following illustrates how I-CAW makes sense of digital traces according to the user’s interest. In this Figure a user browsed interpersonal communication aspects related to the job interview and selected a particular comment of interest. From the content of the comment, I-CAW pulled together a collection of other relevant content, including a video from YouTube with a comment which suggested certain indicators of nervousness observed in the video. From here the user may click on nervousness or another keyword for other content.


alternate textCaption: screenshot from I-CAW - showing a semantically augmented comment and relevant concepts for exploration



Overview Presentation

Intelligent Content Assembly Workbench on Prezi


Demo Screencast of I-CAW



Online Access

I-CAW web application »

Slideshare |