Data mining : past present and future - a typical survey on data streams

📝 Abstract

Data Stream Mining is one of the area gaining lot of practical significance and is progressing at a brisk pace with new methods, methodologies and findings in various applications related to medicine, computer science, bioinformatics and stock market prediction, weather forecast, text, audio and video processing to name a few. Data happens to be the key concern in data mining. With the huge online data generated from several sensors, Internet Relay Chats, Twitter, Face book, Online Bank or ATM Transactions, the concept of dynamically changing data is becoming a key challenge, what we call as data streams. In this paper, we give the algorithm for finding frequent patterns from data streams with a case study and identify the research issues in handling data streams.

💡 Analysis

Data Stream Mining is one of the area gaining lot of practical significance and is progressing at a brisk pace with new methods, methodologies and findings in various applications related to medicine, computer science, bioinformatics and stock market prediction, weather forecast, text, audio and video processing to name a few. Data happens to be the key concern in data mining. With the huge online data generated from several sensors, Internet Relay Chats, Twitter, Face book, Online Bank or ATM Transactions, the concept of dynamically changing data is becoming a key challenge, what we call as data streams. In this paper, we give the algorithm for finding frequent patterns from data streams with a case study and identify the research issues in handling data streams.

📄 Content

Procedia Technology 12 ( 2014 ) 255 – 263 2212-0173 © 2013 The Authors. Published by Elsevier Ltd. Open access under CC BY-NC-ND license. Selection and peer-review under responsibility of the Petru Maior University of Tirgu Mures. doi: 10.1016/j.protcy.2013.12.483 ScienceDirect The 7th International Conference Interdisciplinarity in Engineering (INTER-ENG 2013) Data mining – past, present and future – a typical survey on data streams M.S.B. PhridviRaja,*, C.V. GuruRaob aDepartment of CSE, Kakatiya Institute of Technology and Science, Warangal, INDIA b Department of CSE, S.R. Engineering College (Autonomous), Hasanparthy, Warangal,INDIA Abstract Data Stream Mining is one of the area gaining lot of practical significance and is progressing at a brisk pace with new methods, methodologies and findings in various applications related to medicine, computer science, bioinformatics and stock market prediction, weather forecast, text, audio and video processing to name a few. Data happens to be the key concern in data mining. With the huge online data generated from several sensors, Internet Relay Chats, Twitter, Face book, Online Bank or ATM Transactions, the concept of dynamically changing data is becoming a key challenge, what we call as data streams. In this paper, we give the algorithm for finding frequent patterns from data streams with a case study and identify the research issues in handling data streams. © 2013 The Authors. Published by Elsevier B.V. Selection and peer-review under responsibility of Department of Electrical and Computer Engineering, Faculty of Engineering, “Petru Maior” University of Tîrgu Mureș. Keywords: Clustering; Streams; Mining; Dimensionality reduction; Text stream; Data streams

1. Introduction Data mining is a process of discovering hidden patterns and information from the existing data. The difference between data in the databases and a data warehouse is in a database the data is in the structured form where as in the

• Corresponding author. Tel.: +9030076521. E-mail address: prudviraj.kits@gmail.com Available online at www.sciencedirect.com © 2013 The Authors. Published by Elsevier Ltd. Open access under CC BY-NC-ND license. Selection and peer-review under responsibility of the Petru Maior University of Tirgu Mures. 256
  M.S.B. PhridviRaj and C.V. Guru Rao / Procedia Technology 12 ( 2014 ) 255 – 263 data warehouse the data may or may not be present in the structured format. The structure of the data may be defined to make it compatible for processing. Hence in data mining; we also need to primarily concentrate on cleansing the data so as to make it feasible for further processing. The process of cleansing the data is also called as noise elimination or noise reduction or feature elimination. The process of cleansing data can be either made by using tools such as ETL, tools available in the market or may be done by using various suitable techniques available. The important aspect for consideration in data mining is whether the data considered is static or dynamic. Handling static data is comparatively much easier to handling dynamically varying data. In the case of a static dataset, the entire data is available for analysis purpose in hand before processing and is generally not a time varying data. However dynamic data refers to high voluminous continuously varying information which is not a stand still data and also is not at the hand for processing or analyzing. Data mining requires an algorithm or method to analyze the data of interest. Data may be a sequence data, sequential data, time series, temporal, spatio- temporal, audio signal, video signal to name a few. The concept of data streams has gained a lot of practical interest in the field of data mining. A data stream is an infinite sequence of data points defined usually either using time stamps or an index. We may also view data in the data streams as equivalent to a multidimensional vector containing integer, categorical, graphical with the data in structured or unstructured format. If the data is not structured we may have to transform in to a suitable format for processing by the algorithm being used. With the very high voluminous structured or unstructured continuous data being generated from various applications and devices, the concept of data is no more static but is turning out to be dynamic. This brings a lot of challenges in analyzing the data. Traditional data mining algorithms are not suitable for handling data streams because the algorithms designed perform multiple scans over the data which is not possible when handling the data streams. This brings actual challenge before the data mining researchers working in the area of data streams.
  Further, Many of the existing data mining algorithms available for clustering, classification and finding frequent pattern in the literature are suitable for only static data sets and are no more practically suitable for

This content is AI-processed based on ArXiv data.