Data Mining Research On Text Mining

Abstract - The current classroom utilization of University of Lethbridge is around 50%. Now, they are planning to increase it up to 80%. The data of classrooms for last five years are available that includes Course name, Course Level, Approved size, Sitting types, Actual enrollment and so on. Now, our job is to find out classroom utilization trend, Compare approved and actual enrollment values, find patterns among the classroom size, level and schedule. Above all, based on the data we have to take decision on how we can change the schedule of the classes and their length to have a better utilization yield. We have worked

Text mining is generally identical to content examination it is the way toward getting brilliant data from text. Fantastic data is ordinarily determined through the concocting of examples and patterns through means, for example, factual example learning. Content mining as a rule includes the way toward organizing the information message typically parsing, alongside the expansion of some determined phonetic highlights and the expulsion of others, and ensuing inclusion into a database, inferring designs inside the organized information, lastly assessment and elucidation of the yield. 'High caliber' in content mining more often than not alludes to some blend of pertinence, oddity, and interesting .Text examination programming can help by transposing words and expressions in unstructured information into numerical esteems which would then be able to be connected with organized information in a database and broke down with conventional information mining techniques. Text mining is a minor departure from a field called information mining that tries to discover

Across a wide variety of fields, data emanating from the massive healthcare insurance providers such as government and private companies in healthcare are being collected and stored at tremendous pace. Thus, there is a need felt by most of the companies to manage their wealth of knowledge. Hence, due to the tremendous increase in data, extracting useful information from that data became important. Thus, to extract useful information from the database, Knowledge Discovery in the Database (KDD) is needed. Therefore, KDD is defined as a process of identifying valuable, important, useful and understandable patterns from a large complex database (Maimom & Rokach, 2007).

Text mining is a process which collects information and knowledge from large amounts of unstructured data sources. When I say unstructured data sources, I am talking about Pdf files, Word documents, XML files, text excerpts etc… Text mining collects information from text. Text mining is different than data mining because data mining is a process which collects information and knowledge from large amounts of structured data sources. Structured data sources means that data are classify by categorical, ordinal, or continuous variables, and the goal of data mining is to transform data into model or understandable structure after collecting information from data. However they are

1.3 Literature Review on Seven Data Mining Techniques . . . . . . . . . . . . . . . . . 6

1.1 Text mining: Text mining can be called as text data mining, which is roughly equal to text analytics; text mining is used for deriving high-quality information from text documents and to disclose the unseen meanings. Text mining is more complicated task as compared to data mining because text mining deals with text data, which can be unstructured as well as fuzzy where as data mining is the procedure of extracting information from huge sets of data. Text mining also is called as Text Data Mining (TDM) and Knowledge Discovery in Textual Database (KDT)

The data mining tasks are the kinds of patterns that can be mined. There are many tasks in data mining, the most common ones are: Association, classification, clustering and outlier detections. In the following sections describes the results of applying data mining techniques to the data of our case study for each of the four tasks.

As with the development of the IT technologies , the amount of cumulative data is also Growing. It has resulted large amount of data stock in databases therefore the Data mining comes into model to explore and analyses the databases to extract the interesting and previously obscure patterns and rules well-known as association rule mining It was first introduced in 1993.

Abstract:- Data mining has pulled in a lot of interest in the data business and in the public eye overall lately, because of wide accessibility to massive amount of data and the up and coming need of turning such data into valuable data and information. This data and information can be utilized for applications running from market analysis, fraud detection and different investigations. The aim of this paper is to explore the role of data mining for information extraction, its importance in the current world. prolific writing has been devoted to this research and huge advancements has been made, extending from proficient calculations for continuous thing set and mining in exchange databases to various exploration frontiers, for example, associative classification, correlation mining, pattern based clustering and in addition their wide applications. This paper also presents the scenario about the current issues and challenges data mining is facing.

Data mining is a relatively new phenomenon, therefore the number of peer-reviewed journal articles, blogs and other online sources on the topic are limited but growing rapidly. One key book, Data Mining and Analysis: Fundamental Concepts and Algorithms by Zaki and Meira Jr., takes an algorithmic approach, as the title suggests. Zaki and Meira Jr. define data mining by stating that “data mining comprises the core algorithms that enable one to gain fundamental insights and knowledge

It also helps to present the data with reduced set of samples without representing the whole data set, which reduces the complexity in space and reduction in time. Data mining interesting knowledge includes identifying the relations, differences; groups based one the similar features extracted. Data mining mainly includes the mechanism for representing the data, Specification on required information and method to search the algorithm. Representation model used to represent the underlying data and interpretability of model which interacts with human.

Data mining is the process of extracting knowledge from large data sets. It uses artificial intelligence methods to discover the hidden relationships among the huge amount of data that is collected. It has a great potential to improve applications in many fields like Healthcare systems, Customer relationship management, Financial banking, Research analysis, Bio informatics, Marketing analysis, Education, Manufacturing engineering, Criminology and many more. Criminology is the study of crimes and typically a criminologist’s job include analyzing data to determine why the crime was committed and more importantly to predict and prevent criminal behavior in the future. It became an interesting field to apply data mining techniques because of its large datasets and the complexity of relationships between the data. This paper will discuss some of the tools and techniques used in this field to find out important information that will help and support the police forces and reduce social nuisance.

Abstract—Nowadays, Popularity of Internet and wide improvement in enterprise information is leading to vast research in text and data mining, and information filtering. So, the cluster technology is becoming the core of text mining. Clustering is an important form of data mining. Clustering is a process of grouping similar sets of data into a group, called clusters. This paper comprises of text clustering algorithms, also analysis and comparison of the algorithms are done with respect to the applicable scope, the initial parameters , size of dataset, accuracy, dimensionality, cluster shape and noise sensitivity. Algorithms are classified as partitioned based clustering, hierarchical clustering, density-based , self-organizing maps and fuzzy clustering techniques. The brief idea of each clustering technique is mentioned in this paper.

These necessities have prompted the conception of Data Mining that has been changing the live from the data age toward the coming information age. A considerable amount of literature has been published on Data Mining and the aim of this survey is concerned with the ideas behind the processes; purpose and techniques of Data Mining. [1][2]

From a practical perspective, Data Mining automates the whole process of categorizing and discovering new understandable relationship by using advanced tools and utilizing some basic understanding of statistics, machine learning and database systems. The useful accurate information we acquire after applying this process is reusable and utilized to take important steps towards increased revenue, reduced costs in retail, financial, communication, and marketing business organization. The wide range of applicability in heterogeneous domains which comprises of large volume of rich data makes Data Mining an important and challenging sector for the Data scientists.