ISSN: 2277-9655
[Goudannavar* et al., 7(3): March, 2018] Impact Factor: 5.164
IC™ Value: 3.00 CODEN: IJESS7
http: // www.ijesrt.com© International Journal of Engineering Sciences & Research Technology
[449]
IJESRT
INTERNATIONAL JOURNAL OF ENGINEERING SCIENCES & RESEARCH
TECHNOLOGY
REVIEW ON WEB MULTIMEDIA MINING AND KNOWLEDGE DISCOVERY
Suraj Jain
1
, Siddu P. Algur
2
, *Basavaraj A. Goudannavar
3
, Prashant Bhat
4
1,2,3
Department of Computer Science, Rani Channamma University, Belagavi-591156, Karnataka,
India
4
Department of Business Analytics/Data Science, Chris Institute of Management, Lavasa-412112,
Pune, Maharashtra, India
DOI: 10.5281/zenodo.1199346
ABSTRACT
Web Multimedia data mining (WMDM) can be defined as the process of finding interesting patterns from media
data such as audio, video, image and text that are not ordinarily accessible by basic queries and associated
results. MDM is the mining of knowledge and high level multimedia information from large multimedia
database system. MDM refers to pattern discovery, rule extraction and knowledge acquisition from multimedia
database. To extract knowledge from multimedia database multimedia techniques are used. We compare MDM
techniques with the state of the art data mining techniques involving clustering, classification, sequence pattern
mining, association rule mining and visualization. This paper is a review on Web multimedia mining (WMM)
and Knowledge discovery it elaborates basic concepts, application at various areas, techniques, approaches and
other useful areas which need to be work for WMM. Analyzing this huge amount of multimedia data to discover
useful knowledge is a challenging problem which has opened the opportunity for research in WMM and
knowledge discovery.
KEYWORDS: Multimedia Mining, Feature Extraction, Knowledge discovery
I. INTRODUCTION
The recent abundance of digital information available electronically has made the organization of texture
information into an important task. Web multimedia content mining is burgeoning new technology for
discovering knowledge from web multimedia based on complementary resources.
1.1 Web Multimedia Mining
Knowledge Discovery from Data (KDD) was introduced at the beginning of nineties. Taking into account the
polymorphism of multimedia data, multimedia data mining was recently proposed as a new topic of research
[1].
The World Wide Web serves as huge, widely distributed global information service centre for news,
advertisements, consumer information, financial management, education, government, e-commerce and many
other information services. The size of web is in order of hundreds of terabytes and still growing rapidly. The
properties of web: the huge, diverse, and dynamic and thus raises the scalability, multimedia data and temporal
issues respectively. At the extreme knowledge discovery, i.e., identification of novel information through
inference mechanisms consider as web mining [2]. Historically, the conception of discovering useful patterns in
data has been given a variety of names like data mining, knowledge extraction, Information discovery,
Information harvesting, data archaeology, and data pattern processing.
The Web Mining is about to discovery of knowledge in all its forms, everywhere on the web. It is the
process of discovering potentially useful and previously unknown information or knowledge from the web data.
Three distinct categories based on [3], are: the application of data mining techniques to extract and prepare
knowledge from Web content (include text, image and video), structure (hyperlinks between documents), and
usage (logs of web sites). Web content mining describes the discovery of useful information from web contents
data/documents. The web content consists of several types of data such as textual, image, audio, metadata and
hyper links. The web multimedia contents analysis has been facing lots of research challenges due to the
multimedia mining and knowledge discovery deals with non structured information.