System and Framework for Generating Concept Wise Personalized Books from a Traditional Book
Publication Date: 2014-Jun-19
The IP.com Prior Art Database
Disclosed is a system which given a book and a knowledge graph, generate personalized mini books, each explaining a different concept covered in the book. These mini books are personalized as the generation of the mini books is in accordance with the student’s current knowledge state.
Page 01 of 3
System and Framework for Generating Concept Wise Personalized Books from a
The Education industry is moving towards fine-grained instruction/concept based curriculum design and student learning evaluation (for example, Common Core, Dynamic Learning Maps etc.) which is mainly driven by the demand for Personalized Education.
As part of this transformation, adaptive learning platforms that guide students through different learning paths (sequences of concepts) based on a fine-grained assessment of a student's concept strengths and gaps have immerged. In adaptive systems, the concept-level remedial content is created (e.g. by teacher/SME/platform vendor) and offered through the adaptive platform. The assessments are carried out using questions drawn from large question banks (computer adaptive testing). However, traditional publishers are slow to respond to this need as it requires deep investment in creating large books covering a gamut of topics in a discipline. Also, the dependencies between book chapters/sections/subsections make it difficult to focus on learning an independent concept, and can lead to tedious back-and-forth navigation
Some adaptive learning platforms are trying up with publishers to onboard content onto their
platforms, which is mostly a manual process of content tagging and re-design to suit the needs of adaptive learning.
In the approach for automatic book generation suggested by Chen et al (2005), it takes a topic hierarchy from the user and finds the suitable web pages for all the concepts in the hierarchy. These web pages are selected from documents retrieved by a search engine. Users can later browse through the descriptive web pages.
Another approach for automatically authoring the books has also been proposed by Philip (2007). In this, the system's database is filled with genre-relevant content and specific templates coded to reflect domain knowledge. Based on the interested topic given by the user and the little information about it, the system produces detailed report based on the genre-relevant content stored in the database. The books generated by this approach is of very specific types like technical or business reports, language dictionaries, crossword puzzle book etc.
Page 02 of 3
Figure 1: System Architecture
As shown in Fig.1, 'Segmentation' is used to split a book B into N sequential logical segments. The segmen...