24-09-2016, 02:56 PM
1456117818-5.Report.docx (Size: 217.43 KB / Downloads: 5)
INTRODUCTION
Multimedia data mining is used for extracting interesting information for multimedia data sets, such as audio, video, images, graphics, speech, text and combination of several types of data set which are all converted from different formats into digital media [18]. Multimedia mining is a subfield of data mining which is used to find interesting information of implicit knowledge from multimedia databases. Multimedia data are classified into five types; they are (i) text data (ii) Image data (iii) audio data (iv) video data and (v) electronic and digital ink [2]. Text data can be used in web browsers, messages like MMS and SMS. Image data can be used in art work and pictures with text still images taken by a digital camera. Audio data contains sound, MP3 songs, speech and music. Video data include time aligned sequence of frames, MPEG videos from desktops, cell phones, video cameras [17]. Electronic and digital ink its sequence of time aligned 2D or 3D coordinates of stylus, a light pen, data glove sensors, graphical, similar devices are stored in a multimedia database and use to develop a multimedia system.
Since 1960s the research in the field of multimedia has initiated for combining different multimedia data into one application when text and images were combined in a document. During the research and development process of video synchronization of audio and animation was completed using a timeline to specify when they should be played [2]. The difficulties of multimedia data capture, storage, transmission and presentation have been explored in the middle of 1990s where the multimedia standards MPEG-4, X3D, MPEG-7 and MX have continued to grow. These are reformed and clearly handled sound, images, videos, and 3-D (three-dimension) objects that combined by events, synchronization, scripting languages which describe the content of any multimedia object [5]. For multimedia distribution and database applications different algorithms are required. Such a database can be queried, for example, with the SQL multimedia and application packages known as SQL/MM. Multimedia database system includes a multimedia database management system (MMDBMS) which handles and provides foundation for storing, manipulating and retrieving multimedia data from multimedia database [4]. Multimedia data consists of structured data and unstructured.
Data such as audio, video, graphs, images and text media.
CATEGORIES OF MULTIMEDIA DATA MINING
The multimedia data mining is classified into two broad categories as static media and dynamic media. Static media contains text (digital library, creating SMS and MMS) and images (photos and medical images). Dynamic media contains Audio (music and MP3 sounds) and Video (movies). Multimedia mining refers to analysis of large amount of multimedia information in order to extract patterns based on their statistical relationships. Figure 1 shows the categories of multimedia data mining
2.1 Text mining
Text Mining also referred as text data mining and it is used to find meaningful information from the unstructured texts that are from various sources. Text is the foremost general medium for the proper exchange of information [3]. Text Mining is to evaluate huge amount of usual language text and it detects exact patterns to find useful information.
2.2 Image mining
Image mining systems can discover meaningful information or image patterns from a huge collection of images. Image mining determines how low level pixel representation consists of a raw image or image sequence can be handled to recognize high-level spatial objects and relationship [14]. It includes digital image processing, image understanding, database, AI and so on.
Image mining is the concept used to detect unusual patterns and extract implicit and useful data from images stored in the large data bases. Therefore, we can say that image mining deals with making associations between different images from large image databases. Image mining is used in variety of fields like medical diagnosis, space research, remote sensing, agriculture, industries, and also handling hyper spectral images. Images include maps, geological structures, and biological structures and even in the educational field,
The fundamental challenge in image mining is to reveal out how low-level pixel representation enclosed in a raw image or image sequence can be processed to recognize high-level image objects and relationships.
2.3 Video Mining
Video mining is unsubstantiated to find the interesting patterns from large amount of video data; multimedia data is video data such as text, image, and metadata, visual and audio. The processing are indexing, automatic segmentation, content-based retrieval, classification and detecting triggers. It is commonly used in various applications like security and surveillance, entertainment, medicine, sports and education programs [15].
Mining video data is even more complicated than mining image data. One can regard
video to be a collection of moving images, much like animation. The important areas include developing query and retrieval techniques for video databases, including video indexing, query languages, and optimization strategies.
2.4 Audio mining
Audio mining plays an important role in multimedia applications, is a technique by which the content of an audio signal can be automatically searched, analyzed and rotten with wavelet transformation. Band energy, frequency centroid, zero crossing rate, pitch period and band-width are often used features for audio processing [2]. It is generally used in the field of automatic speech recognition, where the analysis efforts to find any speech within the audio [11].
CHAPTER 3
MULTIMEDIA DATA MINING PROCESS
It shows present architecture which includes the types of multimedia mining process [19]. Data Collection is the initial stage of the learning system; Pre-processing is to extract significant features from raw data, it includes data cleaning, transformation, normalization, feature extraction, etc. Learning can be direct, if informative types can be recognized at pre-processing stage. Complete process depends extremely on the nature of raw data and difficulty’s field. The product of pre-processing is the training set. Specified training set, a learning model has to be selected to learn from it and make multimedia model is more constant.
Converting Un-structured data to structured data:
Data resides in fixed field within a record or file is called structured data and these data are stored in sequential form. Structured data has been easily entered, stored, queried and analyzed. Unstructured data is bit stream, for example pixel representation for an image, audio, video and character representation for text [1]. These sorts of files may have an internal structure, they are still considered “unstructured” because the data they contain does not fit neatly in a database. For example, image and video of different objects has some similarity - each represents an interpretation of a building - but then without clear structure.
Current data mining tool operate on structured data, which resides in huge volume of relational database while data in multimedia databases are semi-structured or un-structured. Hence, the semi-structured or unstructured multimedia data is converted into structured one, and then the current data mining tools are used to extract the knowledge. A difference between unstructured data and structured data mining is the sequence or time element. The architecture of converting unstructured data to structured data and it is used for extracting information from unstructured database is shown in Figure 3. Then data mining tools are applied to the stored structured databases.
CHAPTER 4
ARCHITECTURES FOR MULTIMEDIA DATA MINING
Multimedia mining architecture is given in Figure 4. The architecture has several components. Important components are (1) Input (2) Multimedia Content (3) Spatiotemporal Segmentation (4) Feature Extraction (5) Finding the similar Patterns and (6) Evaluation of Results.
CHAPTER 5
5. MODELS FOR MULTIMEDIA MINING
The models which are used to perform multimedia data are very important in mining. Commonly four different multimedia mining models have been used. These are classification, association rule, clustering and statistical modeling.
5.1 Classification
Classification is a technique for multimedia data analysis, can learn from every property of a specified set of multimedia. It is divided into a predefined class label, so as to achieve the purpose of classification. Classification is the process of constructing data into categories for its better effective and efficient use, it creates a function that well-planned data item into one of many predefined classes, by inputting a training data set and building a model of the class attribute based on the rest of the attributes. Decision tree classification has a perceptive nature that the users conceptual model without loss of exactness. Hidden Markov Model used for classifying the multimedia data such as images and video as indoor-outdoor games [6].
5.2 Association Rule
Association Rule is one of the most important data mining technique which helps to find relations between data items in huge databases. There are two different types of associations in multimedia mining: association between image content and non-image content features [1]. Mining the frequently occurring patterns between different images becomes mining the repeated patterns in a set of transactions. Multi-relational association rule mining is used to display the multiple reports for the same image. In image classification also multiple level association rule techniques are used.
5.3 Clustering
Cluster analysis divides the data objects into multiple groups or clusters. Cluster analysis combines all objects based on their groups. Clustering algorithms can be divided into several methods they are hierarchical methods, density-based methods, grid-based methods, and model-based methods, k-means algorithm and graph based model [3]. In multimedia mining, clustering technique can be applied to group similar images, objects, sounds, videos and texts.
5.4 Statistical Modeling
Statistical mining models are used to regulate the statistical validity of test parameters and have been used to test hypothesis, undertake correlation studies and transform and make data for further analysis. This is used to establish links between words and partitioned image regions to form a simple co-occurrence model