Unit Assessment Type Assessment Number Assessment Name Weighting Alignment with Unit and Course Due Date and Time Group Assignment A4 Data Mining & BI Report 25% ULO1, ULO2, ULO3, ULO4 Assessment...

1 answer below »
Unit Assessment Type Assessment Number Assessment Name Weighting Alignment with Unit and Course Due Date and Time Group Assignment A4 Data Mining & BI Report 25% ULO1, ULO2, ULO3, ULO4 Assessment Description In this assessment, the students will extend their previous work from assessment A3 Business case understanding. Here, the students have to submit a report of the data mining process on a real-world scenario and a presentation and QA Session will be held based on the report written. The report will consist of the details of every step followed by the students. Detailed Submission Requirements Cover Page • Title • Group members Introduction • Importance of the chosen area • Why this data set is interesting • What has been done so far • Which can be done • Description of the present experiment 1. Data preparation and Feature extraction: 1.1 Select data o Task Select data 1.2 Clean data o Task Clean data o Output Data cleaning report 1.3 Construct data/ feature extraction o Task Construct data o Output Derived attributes o Activities: Derived attributes o Add new attributes to the accessed data o Activities Single-attribute transformations o Output Generated records Report (10%): Week 11, Friday, 04 June 2021, 11:59 pm via Moodle. Presentation and QA Session (15%): Week 12 In Class. 2 Modeling 2.1 Select modeling technique o Task – Select Modelling Technique 2.2 Output Modeling technique o Record the actual modeling technique that is used. 2.3 Output Modeling assumption o Activities Define any built-in assumptions made by the technique about the data (e.g. quality, format, distribution). Compare these assumptions with those in the Data Description Report. Make sure that these assumptions hold and step back to the Data Preparation Phase if necessary. You can explain the data file here, even when it is pre prepared. 3 Generate test design 3.1 Task Generate test design o Activities Check existing test designs for each data mining goal separately. Decide on necessary steps (number of iterations, number of folds etc.). Prepare data required for test. (You can use 66% of records for model Building and rest for Testing) 3.2 Build model o Task - Build model Run the modeling tool on the prepared dataset to create one or more models. (Using Knime Tool as shown in the lab). 3.3 Output Parameter settings o Activities - Set initial parameters. Document reasons for choosing those values. o Activities - Run the selected technique on the input dataset to produce the model. Post-process data mining results (e.g. editing rules, display trees). 3.4 Output Model description o Activities - Describe any characteristics of the current model that may be useful for the future. Give a detailed description of the model and any special features. o Activities - State conclusions regarding patterns in the data (if any); sometimes the model reveals important facts about the data without a separate Assessment process (e.g. that the output or conclusion is duplicated in one of the inputs). 4 Evaluation and Conclusion Previous evaluation steps dealt with factors such as the accuracy and generality of the model. This step assesses the degree to which the model meets the business objectives and seeks to determine if there is some business reason why this model is deficient. It compares results with the evaluation criteria defined at the start of the project. A good way of defining the total outputs of a data mining project is to use the equation: RESULTS = MODELS + FINDINGS In this equation we are defining that the total output of the data mining project is not just the models (although they are, of course, important) but also findings which we define as anything (apart from the model) that is important in meeting objectives of the business (or important in leading to new questions, line of approach or side effects (e.g. data quality problems uncovered by the data mining exercise). Note: although the model is directly connected to the business questions, the findings need not be related to any questions or objective, but are important to the initiator of the project. ~ End of Assessment Details ~ Marking Criteria Activities Rank the possible actions. Select one of the possible actions. Document reasons for the choice. Content Marks Cover Page Table of contents 0.5 Executive Summary 0.5 Introduction 0.5 Data Pre-processing and feature extraction 2.5 Experiment 3 Result analysis 2.5 Conclusion 0.5 Presentation and QA 15 Rubrics Marking criteria HD D C P F ULO1: Demonstrate broad understanding of data mining and business intelligence and their benefits to business practice ULO 2: Choose and apply models and key methods for classification, prediction, reduction, exploration, affinity analysis, and customer segmentation that can be applied to data mining as part of a business intelligence strategy ULO3: Analyse appropriate models and methods for classification, prediction, reduction, exploration, affinity analysis, and customer segmentation to data mining ULO4: Propose a data mining approach using real business cases as part of a business intelligence strategy Report, presentation and QA outcome address all the tasks. Report consists of no/minor mistakes. (21-25 marks) Report, presentation and QA outcome address all the tasks. Report consists of a few number of mistakes. (18-20 marks) Report, presentation and QA outcome address most of the contents. Report consists of a few number of mistakes. (15-17 marks) Report, presentation and QA outcome address a few of the contents. Report consists of a good number of mistakes. (13-14 marks) Incomplete report. Unable to perform the experiment/dat a preprocessing/ conclude result. Unable to answer to the question of QA Session and Unable to present the work that has been done. (0-12.5 marks) Misconduct • Engaging someone else to write any part of your assessment for you is classified as misconduct. • To avoid being charged with Misconduct, students need to submit their own work. • Remember that this is a Turnitin assignment and plagiarism will be subject to severe penalties. • The AIH misconduct policy and procedure can be read on the AIH website (https://aih.nsw.edu.au/about-us/policies-procedures/). Late Submission • Late submission is not permitted, practical submission link will close after 1 hour. Special consideration • Students whose ability to submit or attend an assessment item is affected by sickness, misadventure or other circumstances beyond their control, may be eligible for special consideration. No consideration is given when the condition or event is unrelated to the student's performance in a component of the assessment, or when it is considered not to be serious. • Students applying for special consideration must submit the form within 3 days of the due date of the assessment item or exam. • The form can be obtained from the AIH website (https://aih.nsw.edu.au/currentstudents/student-forms/) or on-campus at Reception. • The request form must be submitted to Student Services. Supporting evidence should be attached. For further information please refer to the Student Assessment Policy and associated Procedure available on (https://aih.nsw.edu.au/about-us/policies-procedures/).
Answered 3 days AfterMay 31, 2021BISY3001

Answer To: Unit Assessment Type Assessment Number Assessment Name Weighting Alignment with Unit and Course Due...

Mohd answered on Jun 03 2021
156 Votes
Introduction
In this project we are building predictive models for banking customers churn using Decision tree, random forest and naive bayes classifier algorithms. First we want to find a list of significant contributors to attrition. We have nineteen independent variables(predictors). We have done fe
ature engineering to clean the data. We have chosen a three classification algorithm to estimate the future people's attrition of banking churn data. We have eliminated insignificant and unnecessary predictors from the model. Some variables were eliminated inorder to avoid any kind of multicollinearity presence. Multicollinearity severely affects our model performance and predictive ability. We have partitioned data into two group validation and training. In future several other boosting algorithm models can be applied to this data in order to boost the efficiency of the model.
Importance of the chosen area:
Due to exponential growth in the bank sector. Companies want to keep a close eye on their customers. What are their spending patterns? What are the factors behind particular customers terminating the bank services? We have seen the success of american express, citi bank and other banking services providers. They have built models to classify customers who are likely to attrition, and possible factors behind attrition depending on many significant factors. Better insight into credit card customers to help in making potential decisions regarding credit card services offered by banks.
We can build a prototype using a web page on which we enter several essential information regarding the customer like past transaction info, owner origin and income. We can train models with existing data. Whenever bank officials or executives enter certain required information they could get an insight about their future risk about customer churn.
Why this data set is interesting
This data was retrieved from kaggle, earlier data analysis was done using python and r code . This data consist extensive required features or attributes
What has been done so far:
Earlier researchers have investigated problems related with housing price data to measure the respondent willingness to pay for clean air. They have used hedonic price models and Boston housing data. They have calculated estimates of respondent willingness to pay for clean air.
We are using this data to predict prices using boston data variables,for example crime rate in the area. We can build a linear regression model to estimate house prices for many applications like rental Market real estate and government purposes.
•Description of the present experiment1.
We are using feature engineering, feature selection, and linear regression modelling techniques to build a predictive model of median price for boston houses. We want to assess model performance on validation dataset in order identify whether our model is underfit or overfitted. Both situation must be avoided in order to achieve desired results.
Data preparation and Feature extraction:
Select data:...
SOLUTION.PDF

Answer To This Question Is Available To Download

Related Questions & Answers

More Questions »

Submit New Assignment

Copy and Paste Your Assignment Here