Technology to Empower SDL from the Web
Kirill Kireyev, PhD
The Web has revolutionized almost every aspect of our lives, but the one that is underappreciated is learning. Thanks to the Web, access to educational content is more instantaneous and less expensive than ever before.
Even within a traditional school model, many scholars recognize the Web’s tremendous potential to disrupt one-size-fits-all education towards a more flexible, personalized model. As Milton Chen points out in his book “Education Nation: Six Leading Edges of Innovation in our Schools” (Chen, 2010): “[…] the lecture, like the textbook, is not a very student-friendly way of delivering information. They are one person’s - albeit an expert’s - version of knowledge. Thanks to Google, Bing, and other search engines, students now know there are multiple versions of knowledge about a topic, with various length, depths, points of view and media.”
Even more significant is the Web’s potential to empower lifelong, self-directed learning beyond the traditional classroom. This is particularly important in the today’s knowledge-based economy, where most jobs have extensive, and constantly evolving knowledge prerequisites.
Furthermore, on-demand, self-directed learning is in itself a critical 21st century skill. As professor Michael Eisenberg points out in his column in Seattle Times: “[in order to] go after the high-skill, high-wage jobs of the future […,] we need to add a new "R" — research — to the traditional three Rs of reading, writing and arithmetic. We must also dedicate resources to make this kind of education occur. Students across the country need 21st-century research skills that include abilities to navigate large quantities of information and multiple technologies that deliver it.”
Paradoxically, the large volume of information on the Web made self-directed learning harder, not easier! This is one of the findings of Project Information Literacy, a project that Eisenberg heads up, which studies the challenges that real users face when conducting both school-related and “everyday life” research. Based on examining research habits of thousands of students, Eisenberg and colleagues were able to distill several prominent components (or “contexts”) that learners consistently find laborious when conducting independent research or learning. For example, the “big picture” context involves selecting and defining a topic, understanding multiple sides of an argument, figuring out how the topic might best fit into the course curriculum. This includes understanding what is important to know about a particular topic in the first place, what are the important questions to ask, and where to begin. Another group of researchers (Butcher & Sumner, 2011) refer to this challenge as the "sensemaking paradox":
Self-directed learners face a sensemaking paradox: they must employ deep-level thinking skills in order to process information sources meaningfully, but they often lack the requisite domain knowledge needed to deeply analyze information sources and to successfully integrate incoming information with their own existing knowledge.
A different challenge is the vocabulary or language context, which involves: “[…] becoming more comfortable with the language, terms, and discourse of a topic area. […] Without a context for the vocabulary and terminology of a topic—what things are called and what they mean— students could not proceed with any confidence and reported that they usually did so with little success.” (Project Information Literacy report 2009).
Or, as Butcher (2006) puts it in a different article “subtle differences between concepts can make it difficult for learners without domain expertise to generate relevant search terms.”
instaGrok is an intelligent educational search engine, that is designed to empower self-directed learning from the Web for anyone. It not only finds quality educational information about any topic, but helps the user make sense of it. This is made possible by a number of technical innovations.
First, instaGrok automatically filters out non-educational websites that constitute a large part of typical search results from a general-purpose search engine. For example, a search on “gravity” in Google yields among the top results “Gravity Skateboards”, lyrics for a song called “Gravity” as well as several business homepages. These are clearly irrelevant when one’s aim is to learn about the concept of “gravity”.
In addition, instaGrok automatically analyzes and annotates the (educational) results, and displays important information such as difficulty level (“school” vs “college”), a table of contents, as well as a word cloud corresponding to important concepts covered in each material.
Even more important is the instaGrok’s ability to help the user navigate and make sense of the potentially vast amount of educational information about a topic. On the “Overview” section of the results, instaGrok displays important sub-topics, sentences, and questions, allowing the user to quickly get a taste for the relevant themes (helping with the “big picture” problem). The list of terminology (each of which can be clicked to get a definition) helps the user quickly assess the important vocabulary (aiding with the “vocabulary” problem). In addition, the automatically generated concept graph allows the user to interactively visualize relationships between important concepts.
Another important innovation that instaGrok offers is auto-generated multiple-choice quizzes based on the topic of user’s inquiry. These can be used either for preliminary self-assessment or as a way to review and practice what one has learned. This last point is important – a recent study shows that having an active recall practice as part of learning (such as quizzes or flashcards) significantly improves retention compared to simply passive learning: “Learning is not about studying or getting knowledge ‘in memory’. Learning is about retrieving. So it is important to make retrieval practice an integral part of the learning process.” (Karpicke et al., Science, 2011)
The alpha version of instaGrok is available for free at http://instaGrok.com. We are very interested in connecting with students, teachers and self-directed learners and hearing their feedback as to how we can improve the technology to help their learning and teaching needs.
About the author:
Kirill has a joint PhD in Computer Science and Cognitive Science from the University of Colorado. His graduate research focused on cognitive modeling and building technology for personalized adaptive learning, focusing on vocabulary instruction and motivated by his work at Pearson Education. Kirill is the founder of instaGrok.com, a project that he has been passionately building for the past year and a half.