MS. SUNAYANA GAWDEM.TECH. PART I
14109
MIND MAPPING AND ITS APPLICATIONS,
INTRODUCTION TO CONTEXT TREES
MIND MAP CONCEPT DEFINITION
Mind-mapping is a technique to record and organize information, and to develop new ideas [Holland et al. 2004]
Mind-maps are similar to outlines and consist of three elements, namely nodes, connections, and visual clues.
To begin mind-mapping, users create a root node that represents the central concept that the users are interested in [Davies 2011]. To detail the central concept, users create child-nodes that are connected to the root node. To detail the child-nodes, users create child-nodes for the child-nodes, and so on.
EXAMPLE
MIND MAPS IN HUMAN COMPUTER INTERACTION
Faste and Lin [2012] evaluated the effectiveness of mind- mapping tools and developed a framework for collaboration based on mind-maps.
Document engineering & text mining
Kudelic et al. [2012] created mind-maps from texts automatically.
ANDBia et al. [2010] utilized mind-maps to
model semi-structured documents, i.e. XML files and the corresponding DTDs, schemas, and XML instances.
In the field of education
Jamieson [2012] researched how graph analysis techniques could be used with mind-maps to quantify the learning of students.
ANDSomers et al. [2014] used mind-maps to
research how knowledgeable business school students are.
UTILIZING MIND-MAPS IN IR & USER MODELLING
By Joeran Beel, Stefan Langer, Marcel Genzmehr, Bela Gipp
Published in UMAP 2014Presented 8 ideas on how mind mapping
can be used in IR applicationsUser modelling was the most feasible use
caseProposed to implement a prototype-
Research paper recommender system
ARCHITECTURE OF DOCEAR’S RECOMMENDATION SYSTEM
By Joeran Beel, Stefean Langer, Bela Gipp, Andreas
Published in D-lib magazine of Digital Libraries 2014 AND ACM/IEEE Joint Conference on Digital Libraries 2014
Introduced 4 datasets which contains metadata about research articles, details of Docear’s users and their mind-maps and recommendations they received.
COMPARABILITY OF RECOMMENDER SYSTEM EVALUATIONS AND CHARACTERISTICS OF
DOCEAR’S USERS
By Stefan Langer and Joeran BeelPublished in a workshop: Dimensions and
Design at the ACM RecSys 2014 ConferenceProved that user characteristics affect the
performance of recommender system.
Mind-Map Based User Modelling and Research Paper Recommendations
By Joeran Beel, Stefean Langer, Bela Gipp and Georgia
Published and Presented in UMAP conference 2015
User Models were developed based on unique data from Mind Maps and Recommender system was integrated with Docear.
Raised CTR to 9.82%
Problem Definition
To develop a mini-recommender system
Input from mind maps created by FreeMind
Giving Recommendations from Google based on the content of Mind Maps nodes alone.
Testing
Introducing Context Tree Recommender System
A context-tree recommender system builds a hierarchy of contexts, arranged in a tree
Context can be the list of stories read by a user.Child node completely contains the context of its
parents. The root node corresponds to the most general
context, i.e. when no information is available to profile the user Recommendations on most popular or most recent stories.
More the user browses the stories, the more contexts we are able to extract.
Deeper Context Trees and finer Recommendations.
Example of Context tree
Offline and Online Evaluation of News Recommender Systems at swissinfo.ch
By Florent Garcin, Olivier D, Christophe Bruttin.
Published in ACM RecSys 2014, USA.
CT Recommender System.Profiles the users in real time without Log in.Improves the CTR by 35%
Online CTR with Context Tree
Future Work
CT Recommender System for Audios or Videos
CTR of Recommender Systems:
Standard Method: Up to 3.09% Mind Map Based: Up to 9.82% Context Tree Based: Improved By 35%
What’s Next??
REFERENCES
BEEL, J., LANGER, S., GENZMEHR, M. AND GIP, B., 2014. Utilizing Mind-Maps for Information Retrieval and User Modelling. Proceedings of the 22nd Conference on User Modelling, Adaption, and Personalization (UMAP
BEEL, J., LANGER, S. AND GIPP, B., 2014. The Architecture and Datasets of Docear’s Research Paper Recommender System. In Proceedings of the 3rd International Workshop on Mining Scientific Publications (WOSP 2014) at the ACM/IEEE Joint Conference on Digital Libraries (JCDL 2014).
STEFAN LANGER, BEEL, 2014. Comparability of Recommender System evaluations and characteristics of docear’s users. In ACM RecSys 2014 conference
STEFAN LANGER, BEEL, GIP 2014. Mind-Map Based User Modeling and Research Paper Recommender Systems in ACM Transactions
www.docear.org Florent Garcin, Olivier D, Christophe Bruttin, 2014. Offline and Online
Evaluation of News Recommender Systems at swissinfo.ch in ACM RecSys 2014.
THANK YOU