Data Information Knowledge (KID), Cyclic Process for Retail Business Intelligence

Abstract:

In a highly competitive market, it is very risky for a company to make decision on prediction. They must use their available data. With data analytics we can capture value and benefit from the data. To do this, Company must hire skilled analysts to extract knowledge from the information and they have to make new system to do this task. Company has to invest money and resources. Instead of it, this report presents a KID model based on cognitive approach which can collect data and knowledge by continuously understand data, interpreting data into useful information, taking incoming information and updating knowledge. This approach is applied to a retail business for understanding customer purchasing and product sale situations, so as to support provision of better service and timely adaptation of business strategy.

Introduction:

Data, information, and knowledge are familiar terms in our daily lives. In the information age, data, information and knowledge tend to be used interchangeably. Despite many attempts at their definition and the creation of many relevant definitions, there still seems to be a lack of a clear and complete picture of what these terms mean and how they relate to each other. However, there is a consensus that data, information, and knowledge are part of a sequential order. The definitions of data, information and knowledge and their relations are the premises for building a KID model. We explain them from a pragmatic and implementable view point [1].

Smart business, by definition, indicates the ability to achieve goals which are set according to the development tendency of business. The key to successful implementation of the vision of smart business relies on a comprehensive understanding of the surrounded scenario in which a wide spectrum of elements are concerned. Instances simply include the vision of the company concerned, the global economic situation, current trends, the target market and consumers, etc. It is not difficult to come up with thousands of similar elements for consideration. However, an understanding of customers and products for consumer oriented companies is the important element of data-driven insight [1].

It is believed that big data and advanced analytics can deliver more useful insights than traditional tools. However, this isn’t a given. Companies must capture and manage mountains of data over several years. What is more, they must hire, develop, and retain skilled analysts who can distinguish relevant from irrelevant data, draw the right assumptions, and know what the appropriate tools or algorithms to use for translating information into insights are. The former takes time and the latter increases cost. Moreover, individuals with analytical talent as well as acute business acumen are in high demand and short supply [1].

To lighten the burden on companies and support big data analytics, this paper presents a KID model based cognitive approach. Instead of big data, it continuously perceives incoming data piece-by-piece; interprets them into meaningful information; absorbs information into existing knowledge; and updates knowledge just as humans do. Prior knowledge about customers and products, and expert knowledge and skills in retail marketing can be pre-embedded into a knowledge repository in the KID model. Based on this prior knowledge, experience and new knowledge are continuously accumulated, summarized, and evaluated naturally in the data-information-knowledge cyclic process. To some extent, this knowledge is sufficient to turn data driven insights into effective action on the front line. Meanwhile, it can support big data analytics in distinguishing relevant from irrelevant data, drawing the right assumptions, and translating information into insights [1].

Problem Statement:

It has always been a challenge to transform the available data into useful information and derive specific and timely knowledge about customers, products and markets which in turn can help boost profits, reduce costs and support better and more effective management.

Big data and advanced analytics can deliver more useful insights but this is costly in terms of the time it takes to collect data, requires skilled analysts and dynamic changing customer purchasing behavior.

Instead of skilled analysts required for big data, it is suggested a cognitive model based system be used. When accumulated knowledge is sufficient to form useful insights, it is possible for a cognitive model based system to make reasonable assumptions and select appropriate algorithms, to some extent. A KID model based cognitive approach is proposed to support the process of from data to knowledge and get insight from available data.

Objective:

To counter the shortage of skilled analysts in big data, the KID cognitive model is an alternative or supplemental solution.

Literature Review/Previous Work:

A Generic Formulated KID Model for Pragmatic Processing of Data, Information and Knowledge: This paper is focused on understanding fundamental concepts of data, information, knowledge and their relations, and proposes a generic formulated (knowledge-information-data) KID model for pragmatic transformation processing of data, information, and knowledge. In this paper, the KID model is described with emphasis on formalization of data, information and knowledge and formulation of their interrelations. Three abstract functions, i.e., interpretation(), assimilation() and instantiation() are explained how they pragmatically perform the transformation from data to information and from information to knowledge with a retail business scenario [2].

From Data Mining to Knowledge Discovery in Databases: Data mining and knowledge discovery in databases have been attracting a significant amount of research, industry, and media attention of late. What is all the excitement about? This article provides an overview of this emerging field, clarifying how data mining and knowledge discovery in databases are related both to each other and to related fields, such as machine learning, statistics, and databases. The article mentions particular real-world applications, specific data-mining techniques, challenges involved in real-world applications of knowledge discovery, and current and future research directions in the field [3].

KID Model-Driven Things-Edge-Cloud Computing Paradigm for Traffic Data as a Service: The development of intelligent traffic systems can benefit from the pervasiveness of IoT technologies. In recent years, increasing numbers of devices are connected to the IoT, and new kinds of heterogeneous data sources have been generated. This leads to traffic systems that exist in extended dimensions of data space. Although cloud computing can provide essential services that reduce the computational load on IoT devices, it has its limitations: high network bandwidth consumption, high latency, and high privacy risks. To alleviate these problems, edge computing has emerged to reduce the computational load for achieving TDaaS in a dynamic way. However, how to drive all edge servers' work and meet data service requirements is still a key issue. To address this challenge, this article proposes a novel three-level transparency-of-traffic-data service framework, that is, a KID-driven TEC computing paradigm. Its aim is to enable edge servers to cooperatively work with a cloud server. A case study is presented to demonstrate the feasibility of the proposed new computing paradigm with associated mechanisms. The performance of the proposed system is also compared to other methods [4].

Methodology:

Methodology is a systematic approach with set of rules, procedures and tools used for developing an application. It focuses on analyzing the problems associated with the design and implementation of the product and helps to build solutions with increased efficiency and productivity by following certain methods and principles.

Iterative Method:

Planning & Requirements: As with most any development project, the first step is go through an initial planning stage to map out the specification documents, establish software or hardware requirements, and generally prepare for the upcoming stages of the cycle [5].

Software Requirements

• Programming Language : Python 3.5
• IDE : Anaconda
• Operating System : Windows 10

Hardware Requirements

• Processor : Intel Pentium or above
• RAM : 2GB or above
• Hard Disk : 20GB or above

Analysis & Design: Once planning is complete, an analysis is performed to nail down the appropriate business logic, database models, and the like that will be required at this stage in the project. The design stage also occurs here, establishing any technical requirements (languages, data layers, services, etc) that will be utilized in order to meet the needs of the analysis stage [5]. Databases are Features.csv, sales data-set.csv, stores data-set.csv.

Implementation: With the planning and analysis out of the way, the actual implementation and coding process can now begin. All planning, specification, and design docs up to this point are coded and implemented into this initial iteration of the project [5].

Testing: Testing is done by applying small, medium and large number of values. For all these different cases expected result will be checked [5].

Evaluation: Once all prior stages have been completed, it is time for a thorough evaluation of development up to this stage.

Anticipated Result: The amount of Knowledge and Information we get from the data by applying KID model.

Significance of Study: Without using big data analytics and by spending less resources and money, we will be able to get Knowledge which can be used to increase business profits [5].

Scope of the study: KID model can be used by retail business industry or small and mid-level business.

Future Work: The proposed KID model is not fixed. It is an abstract model. It need to be verified by taking different cases and scenarios.

References

[1] Atsushi Sato and Runhe Huang “From Data to Knowledge: a Cognitive Approach to Retail Business Intelligence”, Hosei University Tokyo, Japan

[2] Atsushi Sato and Runhe Huang “A Generic Formulated KID Model for Pragmatic Processing of Data, Information and Knowledge”, Hosei University Tokyo, Japan

[3] Usama Fayyad, Gregory Piatetsky-Shapiro, and Padhraic Smyth “From Data Mining to Knowledge Discovery in Databases”,

[4] Bowen Du, Runhe Huang, Zhipu Xie, Jianhua Ma, and Weifeng Lv “KID Model-Driven Things-Edge-Cloud Computing Paradigm for Traffic Data as a Service”, Hosei University

[5] Andrew Powell-Morse, December 15, 2016. [Online]. Available: https://airbrake.io/blog/sdlc/iterative-model [Accessed September 12, 2018]

[6] Lego Views. [Online]. Available: https://legoviews.com/2013/04/06/put-knowledge-into-action-and-enhance-organisational-wisdom-lsp-and-dikw/ [Accessed September 12, 2018]

Cite this page