A SURVEY ON FEATURE SELECTION FOR EFFICIENT ECONOMIC BIG DATA ANALYTICS

Ms. DeepthiMogaparthi, Prof. Priyadarshani Kalokhe, Ms. Punam Patil, Ms. Pooja Shedutkar, Ms. Sharda Tenginkai. Computer Department Alard College Of Engineering & Management, Pune. ...................................................................................................................... Manuscript Info Abstract ......................... ........................................................................ Manuscript History

Huge amount of data gets collected every day and there is also a need of technology to handle enormous amount of economic data. Hence there are various and huge number of opportunities for economic analysis.Lowquality, high-dimensionality and great volume pose great challenges on efficient analysis of economic big data. To overcome these challenges our paper presents a new structure for efficient analysis of high-dimensional economic big data based on innovative distributed feature selection. The presented framework combines the methods of economic feature selection and econometric model construction to discover the hidden patterns for economic development.
When consumers purchase products through online, products information such as ratings, product reviews, product descriptions given by sellers are very useful for consumers to optimize their purchasing decisions. However, when a consumer purchases used products via online e-commerce sites, the consumer may consider much more attributes about the products than that for purchasing new products. This is due to the need for understanding instance-specific conditions before purchasing a used product and thus the available descriptions for a used product may differ in each other. [20] Big Data analytics requires business processes to change and it must align with the organization's IT infrastructure to support the business initiatives. New ways of doing data analytics and business intelligence impact on technology infrastructure components.
Here we explore the hidden relations between economy and its response indicators from a new angle and extract the meaningful knowledge from economic big data in order to derive right insights and conclusions on an innovative distributed feature selection that integrates advanced feature selection techniques and econometric methods.

ISSN: 2320-5407
Int. J. Adv. Res. 4(11), 1351-1355 1354 business sector and also had to compare these two sectors in forms of goals, missions, decision-making processes, decision actors and organizational structure.

Existing system:-
With promptly increasing popularity of economic activities, large number of records are involved in economic development. [20] Existing system involves limited indicator and requires prior knowledge of economist . [16] When compressing large varieties of economic data, existing methods gives unsatisfactory performance.There are some issues of economic data which pose great challenges:(1)The collected large amount of data contains incorrect,missing values and nonstandard items.(2)The huge dimensionality of economic indicators makes manual feature selection for economic model construction impossible.(3 ) [10] Statistical analysis software frequently generates runtime errors when dealing with the high-dimensionality and huge volume economic data.

Proposed system:-
We present a novel framework combining distributed feature selection methodsand econometric models for efficient economic analysis, which can reveal the valuable insights from the low-quality, high dimensionality, [1] and huge volume economic big data. We develop a subtractive clustering based feature selection algorithm and [6] an attribute coordination based clustering algorithm to select and identify the important features of data in horizontally and vertically. Also, we extend these two methods to distributed platform for economic big data analysis. We conduct correlative and collaborative analysis simultaneously to explore the direct and indirect relations between economy and its response indicators based on the identified economic features. We evaluate the proposed framework and algorithms on the economic development data. [16] Extensive experiments and analysis demonstrate that the designed framework and algorithms can distil the hidden patterns of economic development efficiently and the achieved results accord with the actual development situation.

Conclusion:-
In our paper, we reviewed studies on the data analytics from the traditional data analytics to the recent big data analysis based on economic data. Herewe put forth a hardback feature selection based technology. [16] In order to reduce the noise and for the promotion of the data quality various techniques are approached for cleaning and transforming the collected economic big data. [1] Greatchallenges come in site for a user to quickly measurethe opinions when the UI is impressed by a huge amount of information. Our comprehensive experiments on both discrete and continuous data sets and multiple types of classifiers demonstrate the classification accuracy. [20] Organizing the large amount of messages and information into clusters with meaningful cluster tags or labels, provides an sketch of the content to fulfil users' information or data needs.
Combining the power of Big Data we extract the benefits of Big Data in the e-commercial sites. Various ways of doing economic or commercial data analytics and business intelligence provide an impact on technology framework components. Industries must focus on this now so that they can gain competitive advantage in the market place.