K. M. Yatheendra Pravan, Udhayakumar Shanmugam, P. Rajaraman
{"title":"INSIGHTS! - a modern deep learning approach to data analysis using Feature Name Substitution Network","authors":"K. M. Yatheendra Pravan, Udhayakumar Shanmugam, P. Rajaraman","doi":"10.1109/ICCIDS.2019.8862071","DOIUrl":null,"url":null,"abstract":"The core of technological advancements in the current trend is based on the manipulation of the inestimable amount of data that is generated every second around us. Gaining interesting insights from the data is of utmost importance and the need of the hour. The proposed system makes use of advancements in the domain of deep learning by implementing various algorithms and methodologies to automate the process of data analytics. The intended insights platform is developed using various deep learning frameworks such as Tensorflow, Keras and delivered to the end user as a web platform using Django Framework. The underlying algorithm of insights which makes the automation of analytics possible relies on the efficacy of feature name substitution network implemented using LSTM and the enhanced correlation analysis. These are then used to determine a measure called Insight Relevance Index (IRI) which then updates the global rule set records in the centralized data store accordingly. Employing the proposed system will definitely aid the profit and future growth of an institution or an organization.","PeriodicalId":196915,"journal":{"name":"2019 International Conference on Computational Intelligence in Data Science (ICCIDS)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2019-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2019 International Conference on Computational Intelligence in Data Science (ICCIDS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCIDS.2019.8862071","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
The core of technological advancements in the current trend is based on the manipulation of the inestimable amount of data that is generated every second around us. Gaining interesting insights from the data is of utmost importance and the need of the hour. The proposed system makes use of advancements in the domain of deep learning by implementing various algorithms and methodologies to automate the process of data analytics. The intended insights platform is developed using various deep learning frameworks such as Tensorflow, Keras and delivered to the end user as a web platform using Django Framework. The underlying algorithm of insights which makes the automation of analytics possible relies on the efficacy of feature name substitution network implemented using LSTM and the enhanced correlation analysis. These are then used to determine a measure called Insight Relevance Index (IRI) which then updates the global rule set records in the centralized data store accordingly. Employing the proposed system will definitely aid the profit and future growth of an institution or an organization.
在当前的趋势中,技术进步的核心是基于对我们周围每秒产生的不可估量的数据的操纵。从数据中获得有趣的见解是至关重要的,也是当前的需要。提出的系统通过实现各种算法和方法来自动化数据分析过程,利用深度学习领域的进步。预期的洞察平台是使用各种深度学习框架(如Tensorflow, Keras)开发的,并使用Django框架作为web平台交付给最终用户。使分析自动化成为可能的底层洞察算法依赖于使用LSTM实现的特征名称替换网络的有效性和增强的相关性分析。然后使用这些数据来确定称为Insight Relevance Index (IRI)的度量,IRI随后相应地更新集中式数据存储中的全局规则集记录。采用拟议的系统肯定会有助于一个机构或组织的利润和未来的增长。