Retrieval and extraction of unique patterns from compressed text data using the SVD technique on Hadoop Apache MAHOUT framework

Nowadays Designing of an efficient information retrieval system for multimedia dataset is a big task to understand the trend. Searching for the repeated pattern within a particular genetic sequence has become a much required task in the data mining sectors. Due to the increasing size of text and audio data over internet, various techniques are needed to help with the finding and extraction of very specific information relevant to a user's task or finding the new trends. Matching the unique patterns and generate the rules by using different Pattern Matching, Rule Generation Algorithm. In addition to extracting questionnaires and curiosity based sentences patterns from large database some different implementation of the pattern matching algorithms is proposed. Real world data is large in size like images, speech signals holding high dimensions to represent data. Multiple dimensional data are more critical for detecting and developing the associations among terms. Dimensionality reduction is a technique used for reducing complexity and gives the most frequent item from high dimensional data. It reduces the dimensions of the original input data. Singular value decomposition this dimensionality reduction technique is used for large data reduction on Apache MAHOUT of Hadoop framework. Finally, retrieve and extract the curiosity and questionnaires based unique pattern from large data size using the Map Reduce Framework and compress the result using the SVD technique to give curiosity and questionnaires based subject.