Sales Prediction In Tourism Industry

downloadDownload
  • Words 1908
  • Pages 4
Download PDF

Abstract— The present forceful circumstance for Indian travel and the tourism industry is the need to raise their market and keep control their organizations to utilize data mining devices and methods to create, advertise the travel industry items and administrations. Despite the fact that there are many anticipating models for deciding deals in the tourism industry, data mining systems have been viewed as the best method for determining deals in the travel industry. Data mining is characterized as the way toward discovering valuable examples, connections, and rules, which are not known beforehand, by sifting through a lot of information put away in some repositories(database).

Deals designs from sales information demonstrate advertise slants and can be utilized in anticipating which has incredible potential for dynamic, vital arranging and market rivalry. In this project we use multiple linear regression (MLR) technique to predict sales from historical data and get profit in tourism industry. So, this project aims that how to increase sales from tourism industry using data mining.

Click to get a unique essay

Our writers can write you a new plagiarism-free essay on any topic

Keywords: Prediction, Multiple linear regression, Travel and Tourism industry, data mining.

I. Introduction

In recent few year tourism industry become very important and main incoming source for India The travel industry in India is significant for the nation’s economy and is developing quickly. The World Travel and Tourism Council determined that travel industry produced ₹16.91 lakh crore (US$240 billion) or 9.2% of India’s GDP in 2018 and bolstered 42.673 million occupations, 8.1% of its absolute work. The area is anticipated to develop at a yearly pace of 6.9% to ₹32.05 lakh crore (US$450 billion) by 2028 (9.9% of GDP)[2].

Sales prediction is the way toward assessing the amount of an item or administration that shoppers will buy. Request estimating fundamentally includes strategies including both casual techniques as conjectures, and quantitative strategies, for example, the utilization of recorded deals information or current information from test markets. Sales prediction might be utilized in settling on evaluating choices, in surveying future limit necessities, or in settling on choices on whether to enter another market. The estimate of the travel industry of uncommon significance since it is a marker of future interest, along these lines giving fundamental data to ensuing arranging and strategy making. The main prediction methods which are used in the tourism and recreation fields are time-series methods, multiple regression methods, multivariate methods and qualitative forecasting methods.[4]

II. Literature Survey

In earlier studies, there are various techniques and methods can be used to predict sales and get accurate result.Some of them are can be reffered by us during production of our project which can be explain in following section. These methods are used to give the accurate sales prediction model . These methods are going to give different result on different data set and in different situation. As technology emerging very rapidly, different methods and different data mining algorithms are tested. Due to increase in competition in tourism industry and growth of tourism in India many researcher are take in interest in these topic.

Aditya Joshi, Nidhi Pandey, (Professor) Rashmi Chawla, Pratik Patil are proposed a paper in which they used clustering association mining methods on stock data to classify and find association pattern on sale. From their experimental result these method is very efficient for large stock data and predicting factor of sale. But their method is very simple. Limitation of study is, it requires proper formatted data with specific attribute.[1]

Shini Renjith , A. Sreekumar , M. Jathavedan proposed a paper Evaluation of Partitioning Clustering Algorithms for Processing Social Media Data in Tourism Domain in they used different data mining algorithms to analyse and classify accuracy of that on predicting sales on destination reviews across South India published on holidayiq.com data set .they have used k-means ,k-mediod , CLARA algorithms for predicting sales of given data set.[7]

Girish Kumar Sharma , Promila Sharma proposed a paper A Study of Data Mining Algorithm for Tourism Industry in which they used different data mining algorithm such as decision tree ,classification, clustering, association rule to access a pattern in tourism industry. This paper has overall goal of use different data mining algorithm for tourism industry.[2]

III. Project Scope

The main scope of our project is to increase sales of tourism using different data mining techniques according to its prediction upon historical or collected data set from different tourism website. Using our project tourism administrator can increase their sales. Because, using collected data and prediction on it gives idea about which packages are going to put to customer in different season and different location.

As competition in tourism industry increases customer satisfaction is very important to get increase in sales of tourism. Sales Prediction in tourism has a huge scope for the generations to come.

IV. Proposed system

In this chapter we have give the brief description of proposed system and examine different module of our system along with different model models through which this system is understood and represented. Some of the techniques of data mining are also examine and study. Accordingly we implement our system.

1. Data Mining Techniques

For our project, we studied a many different data mining algorithm such as k-mean, k-mediod , ANN, MFP, etc. But we have select MLR means multiple linear regression algorithm because it has higher accuracy than other algorithm as studied. Multiple linear regression (MLR), also known simply as multiple regression, is a statistical technique that uses several explanatory variables to predict the outcome of a response variable. The goal of multiple linear regression (MLR) is to model the linear relationship between the explanatory (independent) variables and response (dependent) variable. It can be used for sales prediction on the bases of our different attribute rating and cost in tourism data set. According to these techniques and prediction get from our model Admin can add and remove different packages in our system to increase their sales in upcoming season.

2. Hardware And Software Required

2.1. Hardware Requirements

The solution makes extensive use of hardware device including:

  • Processor – i3
  • Hard Disk – 5 GB
  • Memory – 1GB RAM

2.2. Software Requirements

The solution makes extensive use of software including:

  • Windows 7 or higher.
  • Python

3. Python Libraries

For prediction of sales we use python language as it is very effective in use for these type of system. In python there are many libraries for such prediction model. We have used some of which are as follow: pandas ,Jupyter, Jupyter notebook, Numpy, Scikit-learn etc.

4. Data Flow Diagram

A data flow diagram is one of the most common methods used for graphically representing a data through an information system. A DFD shows what information is provided as input to and output from the system.

Figure 3.1 DFD Level 0

Above Figure level 0 also known as Context level DFD, gives the overview of whole system in which the external entities like the tourism data and data mining techniques are process, which are processed for predicting the sales.

Figure 3.2 DFD Level 1

It is elaboration of level 0 DFD. Pre-processing method is done in traditional way .After the-processing the dataset is applied to the data mining algorithm MLR.

V. Methodology

In our system various module are available for testing, evaluating, data set for processing and predicting sales.

Prediction Or Admin Module

This module is our main focus of project where we have predicting module as well. According to our database and historical data set of the sales, system will predict sales for that season of tourism .In recent year tourism become very important aspect in India and competition is very high .So ,to attract customer and give good service becomes essential part. This prediction will help in letting the admin know that which package should remove and add in which period of time or season.

User Module

User module has simple system as every tourism website. This module contain information about tour packages and cost of them. In that module user can enquire about different tour packages and book for that packages with secure payment system. User also cancel their booking for any reason in few hours. Customer also give feedback about its experience and also give rating for that to give suggestion to other customer.

Data Set

In order to get data set we create data set up to 200 which contain different attribute for prediction using MLR algorithm.

Such attribute are: Place, Customer name , Rating , Cost.

Rating can be given in five different way such as:

  1. Poor
  2. Less poor
  3. Moderate
  4. Good
  5. Excellent.

VI. Results

This chapter contain some view of our system along with the result of our work. It contain output of our system which we are studied in this paper and experience by different user.

1. Snapshot Of Project

In first stage of project we have create a data set by processing it. Fig 4.1 represent that first data should be tested and than it was given for training to predict sales from that data set. Training of data is represented in FIG. 4.2 as follow.

Figure 4.1 Command for processing created data-set

Figure 4.2 shows algorithm to predict

After that in second stage, our system GUI is implemented to interact with Admin and customer of tourism website.

Fig. 4.3 shows that how admin can do their work and such like add or remove packages in different season using prediction.

Figure 4.3 Admin module

As shown in Fig. 4.4 according to prediction module graph are display to add or remove packages by admin to give correct choice to customer and grow their sales during that season.

Figure 4.4 represent graph for sales prediction

VII. Conclusion

In this project, we studied different data mining algorithms for sales prediction in tourism industry. According to our study we select best algorithm that is MLR which can predict sales accurately than other algorithm. MLR technique is suitable for our project because it is very efficient and accurate for large data set as for our project. On the bases of our assumption we create efficient and accurate system for predicting sales and according to that add and remove packages to increase their sales for in future. As our main project focus is for tourism industry admin that how they can increase their sales or growth in these competitive environment. It also has some limitation in this project that we cannot used sentiment analysis through which user verbal feedback is not analyse. Finally, according to our paper we successfully created predictive module from MLR that predict any sales in tourism to increase sales.

References

  1. Aditya Joshi, Nidhi Pandey, (Professor) Rashmi Chawla, Pratik Patil. ” Use of Data Mining Techniques to Improve the Effectiveness of Sales and Marketing” IJCSMC, Vol. 4, Issue. 4, April 2015, pg.81 – 87.
  2. Girish Kumar Sharma , Promila Sharma. “A Study of Data Mining Algorithm for Tourism Industry” International Journal of Latest Trends in Engineering and Technology (IJLTET), Vol 7 issue 1 May 2016.
  3. Manoj B. Kathariya , Ravinder Singh Sakshi , Diler Singh Sakshi , Dhaval R. Kathariya. “Data Mining for Travels and Tourism”, Journal of Information and Operations Management, , Volume 3, Issue 1, 2012, pp-114-118, 31 May 2014.
  4. Shini Renjith1 , Anjali C2. ” A Personalized Travel Recommender Model Based on Content-based Prediction and Collaborative Recommendation” , International Journal of Computer Science and Mobile Computing, ICMIC13, December- 2013, pg. 66-73
  5. Mirjana Pejic Bach , Markus Schatten , Zrinka Marušic. “Data Mining Applications in Tourism: A Keyword Analysis” Central European Conference on Information and Intelligent Systems , September 18-20, 2013.
  6. Shini Renjith, Anjali C. “A Personalized Mobile Travel Recommender System using Hybrid Algorithm” 2014 First International Conference on Computational Systems and Communications (ICCSC) , 17-18 December 2014.
  7. Shini Renjith , A. Sreekumar , M. Jathavedan. “Evaluation of Partitioning Clustering Algorithms for Processing Social Media Data in Tourism Domain” 2018 IEEE Recent Advances in Intelligent Computational Systems (RAICS) , December 06 – 08, 2018.

image

We use cookies to give you the best experience possible. By continuing we’ll assume you board with our cookie policy.