Data Mining in Chain Hotels - Assignment Example

Data Mining in Chain Hotels Executive Summary The scenario provided needs to undertake research for data mining in Hotel industry to help the management make decisions in order to maximize profits from their accommodation facilities. Data mining techniques uses the advance in technology to help users generate data, analyze and give solutions to problems in this sector. Using the techniques of data mining, it is possible to come up with a model for data visualization used to present the information to the client easily. This study seeks to help the hotel industry develop a database for many of its operations. Databases can be used by several users seeking businesses in this sector. It helps them to overcome challenges of competition and meet the demands in the market. Problem Statement This study seeks to develop a database for hotel chain management operating 20 hotels in 4 countries. The data mining for the store of information for each hotel and performs analysis with regard to the given hotel. For each hotel the data warehouse will store its name, type, address, country, region, postcode, phone number, and name of the manager. The data also include different types of rooms like single, double, family, suits, etc. Each room may also incorporate certain optional features, such as refrigerator, kitchenette, or laundry. The system should have each room described as room’s type, size, number of beds, maximum number of customers, refrigerator (Boolean), kitchenette (Boolean), laundry (Boolean). The capacity of the hotel chain to accommodate customers is limited. The database should help the management on how to price the hotel rooms in order to realize maximum revenue collection. Looking at the capacity of the hotel over time given in the data warehouse, they can easily come up with the prices. Comparison between the occupancy rate (utilization) and the vacancy rate is considered. Problem Justification The hotel chain’s capacity to accommodate customers is limited. Each hotel has a set number of rooms. The primary source of revenue is accommodation in hotel rooms. The biggest challenge the company faces is determining how to price the hotel rooms. If they are priced low, the hotels will be constantly booked and therefore customers will be forced to try other hotels in competition with The Grande Chat and if the rooms are priced too high, a lot of rooms will remain empty. The hotel chain management wants to realize profits. The only way is to use the data mining to realize their underlying, interesting patterns and relationships that lie hidden within the analysis (Data mining). The expertise is needed in so as to generate the data warehouse which is responsible for fetching relevant information based on the user’s data mining request and the knowledge base used to guide the research. The problem is to ensure that rules are updated and the information is consistent. In hotel industry, knowing where your guests come from, how much they spend on what can help in formulating strategies and maximizing profits. Identifying relationships and variables in the consumer-information systems can be easily done using the data mining (Fayyad, Shapiro & Smyth, 1996). Assumptions made in this scenario is that; the chain hotel has constant number of rules which do not change and also the data is up-to-date. Introduction Data mining is the method of discovering interesting patterns and relationships that are hidden within the large databases. Also it is the analysis of observable data sets in order to find unsuspected relations and give data summary in novel ways that are understandable and also useful to data owner. Data used to identify problems on regular basis (every two weeks) before they become difficult for functional solutions to be reached. Investigation and Scope The data collected from the hotel management which include the following details of the rooms in the hotel chain; name, type and address, country, region, postcode, phone number, and name of the manager. Other information about the rooms description include; room’s type, size, number of beds, maximum number of customers, refrigerator, kitchenette, laundry. The scope of the scenario is within the hospitality industry. Data can also be sourced externally from demographics and websites. Just to build a knowledge base for the company’s management to realize its goals and objectives. The data found is fed into data mining and analyses made on the basis of the amount of data provided. Most data-based modeling is done for specific application domain. Raw data sets initially prepared in data mining are large, related to humans with the potential of being messy. Real databases in the world are subjected to inconsistent data because they of their huge size. Plan for the report and Schedule of tasks The hotel chain data is taken into the data mining process. The process has four parts: 1) selecting data mining task, 2) selecting the method of the data mining, 3) select the suitable algorithm and, 4) extracting the knowledge. Data mining task involves clustering, association rules, summation and sequence discovery depending on whether the model is descriptive or predictive. Methods used can be Rule Induction, Decision trees, or Fuzzy logic. Construct specific algorithm so as to implement general rules which has model presentation, model evaluation and the search. Finally, the answers to the problem solved will result after the simulation for specific algorithm. The processed data will give the report for the scenario. The data mining techniques include; 1) Database oriented approaches. Paralleling the popularity of the data mining itself in the development of new techniques is exploding as well, 2) Statistical approaches. Most innovations are vendor-specific which sometimes does little in advancing the state of the art, 3) Machine learning approaches which provide for the suitable model that matches the data being processed. Regardless of the investigation, data-mining techniques always tend to fall into four major categories: 1) Classification, 2) Association, 3) Sequencing and 4) Clustering. Therefore, most methods begin with a model of the task for which the Expert System is built. A description of the domain is generated in addition to the model. Finally, the knowledge engineer uses hypotheses, models and cognitive analysis techniques to elicit the problem-solving knowledge from the experts. These techniques are also called knowledge data discovery (KDD), and include statistical analysis, neural or fuzzy logic, data visualization or intelligent agents. Data visualization is the presentation of data mining in a beautiful, descriptive and elegant way. The aim is to make the user built a knowledge base of what is going on. There are various ways to visualize data- tables, pie charts, histograms and bar charts. For the scenario, I would use tables to show the occupancy rates and vacancy rates. The model used for the data mining technique would be given in the tables. A report of the breakdown of the revenue from accommodation in the database is straight forward for the management to understand because they intuitively know its existence. Visualization should be used to maximize value to the viewer (Magnini et al., 2003). In data mining users should be skilled because they need to supply correct data and to make objective conclusions, it needs a specialist. This becomes a limitation if the user gives the wrong information, the outcome is also affected. Furthermore, the data mining cannot promise perfect results, cannot correct problems in the data and cannot explain why the outcome occurs. The circumstance which the data mining would fail would be when an incorrect data is given. The challenge is to harness and use the power of technology to design a reliable, decision-making and easy to use process. As the data amounts increase, data mining becomes the tool to be used in optimizing revenue. De-identified data is used for protection by the company which may be mandated by ethical guidelines or legislation governing research. The data can be de-identified when it is shared, published or re-used (Brunk, Kelly & Kohavi, 1997). Design A de-identified data set may contain individual’s age, relatives or household members. Names, addresses, date of birth, gender or other identifying information may be removed from datasets, encrypted or coded. Information can also be masked by changing values or aggregation. The set of data having de-identified data such as names, any geo-codes to identify an individual’s household such as Post Office Box Number or Street Address, telephone numbers, Fax numbers, Electronic mail addresses, Social Security Numbers, Account numbers, Medical record numbers and Health plan beneficiary identifiers, Certificate or license numbers, serial numbers and vehicle identifiers, license plate numbers, medical device identifiers and serial numbers, web universal resource locators (URL), Internet Protocol address (IP) numbers, Biometric identifiers, including finger and voice prints, full face photographic images Hoteliers need to establish which data is to be analyzed. They may send web crawlers to user-defined set of URL (Chen, Han &Yu, 1996). Table 1: Rates of hotels RATES NAME COUNTRY SINGLE DOUBLE FAMILY NUMBER OF VISITORS NO, OF ROOMS HOTEL A AUSTRALIA USD 200 USD 250 USD 300 800 800 HOTEL B KENYA USD 110 USD 150 USD 200 400 500 HOTEL C USA USD 240 USD 265 USD 310 1500 2000 HOTEL D S. AFRICA USD 160 USD 180 USD 250 1000 1500 HOTEL E UK USD 240 USD 270 USD350 1600 2000 From the tables the rates vary in different countries. It was also found that guests prefer rooms with additional features. The table shows that the vacancy rate is high if the rates are high. Data mining software’s available for data processing and analysis; PRODUCT NAME VENDOR URL Intelligent Miner for Text IBM http://www-3.ibm.com SAS Text Mine SAS www.sas.com/datamining ExactAnswer InsightSoft-M www.insight.com.ru/products SPSS Lexi Quest Mine SPSS www.spss.com Online Miner Temis Group www.temis-group.com/ MindServer Recommind Inc. www.recommind.com/english/solutions/ Spy-EM University of Illinois www.cs.uic.edu/ Management Technology www.readware.com/ Xanalys Indexer Xanalys www.xanalys.com/solutions/ Conclusion Data mining is defined as the process of generating interesting patterns from large databases. It is the solution to data analysis problems facing many organizations. Data mining is still in its infancy and so a lot of work is required in data mining research and development. Data mining models generates results known to the user earlier. It is therefore important that visualization should be provide sufficient understanding and trust to the user. Data mining is an alternative to human analysis of information. It reduces human labor effectively in identification, storage and business analysis. The study uses the data mining to advise the chain hotel managers in order to maximize revenue. It helps to translate online data into meaningful competition and customer intelligence used in managerial decision making. Hoteliers need to pay attention to data mining. Although data mining has significant business implications, it is important to acknowledge the techniques as well as the uncertainty and complexity of the future technological developments. This process is however limited to skilled people who can give correct data for the intended output. References Brunk, C., Kelly, A. & Kohavi, R. (1997). MineSet: An Integrated System for Data Access, Visual Data Mining, and Analytical Data Mining. Proceedings of the Third Conference on Knowledge Discovery and Data Mining (KDD-97), Newport Beach, CA. Chen, M.,Han, J. & Yu, P. (1996). Data mining: An overview from a database perspective: IEEE Transactions on knowledge Discovery and Data mining, AAAI. London: MIT Press. Fayyad, U. M.,Shapiro, P. & Smyth, K. ( 1996). From Data Mining to Knowledge Discovery in Databases. AI Magazine, 17(3): 37-54. Magnini, V., Earl, D. Honeycutt Jr. & Sharon K. (2003). Data Mining for Hotel Firms: Use and Limitations. Cornell Hotel and Restaurant Administration Quarterly, 22, 51-61. Ranjan. J. (2009). Data Mining in Pharma Sector : Benefit. Ghaziabad. India: Emerald. Read More