Dss On Life Insurance

DSS ON LIFE INSURANCE
Mining insurance plans for extracting useful patterns and guiding customers in taking the right decision
Introduction to Problem
Life Insurance Investment is one of the most important decisions of ones life and hence decisions from what kind of insurance is needed, what should be the term of insurance to what payments should be made on a regular interval and till when, are the factors that the customer has to zero in on. These decisions can not be made on some insurance agents suggestions, it needs a lot of research work for the customer and also a prior knowledge of the different aspects of an insurance plan.
Motivation
Neural networks have seen an explosion of interest over the last few years and are being successfully applied across an extraordinary range of problem domains, in areas as diverse as finance, medicine, engineering, geology, physics and biology. The excitement stems from the fact that these networks are attempts to model the capabilities of the human brain. Application of Artificial Neural Networks in the field of life insurance will yield immediate automated results, increased productivity from less knowledgeable workers and customer
Revised Problem Statement

Application of Artificial Neural Networks in the field of life insurance to provide immediate, completely reliable and the most appropriate automated results to the customers.
Literature Survey
Artificial neural networks (ANNs) are non-linear data driven self adaptive approach as opposed to the traditional model based methods. They are powerful tools for modelling, especially when the underlying data relationship is unknown. They can identify and learn correlated patterns between input data sets and corresponding target values. After training, ANNs can be used to predict the outcome of new independent input data. They imitate the learning process of the human brain and can process problems involving non-linear and complex data even if the data are imprecise and noisy. Thus they are ideally suited for the modeling of data which are known to be
Characteristics of neural networks Exhibit mapping capabilities Learn by examples Possess the capability to generalize. Are robust systems and are fault tolerant. Can process information in parallel, at high speed, and in a distributed manner. Types of Neural Networks Feed Forward Neural Networks Recurrent Networks
Important concepts used..
Experimenting On WEKA Tool
Pre-process: Choose and modify the data being acted on. Classify: Train and test learning schemes that classify or perform regression. Cluster: Learn clusters for the data. Associate: Learn association rules for the data. Select attribute: Select the most relevant attributes in the data. Visualize: View an interactive 2D plot of the data.
Development of an ANN model

The various steps in developing a neural network model are Variable selection The input variables important for modeling variable(s) under study are selected by suitable variable selection procedures.
Formation of training, testing and validation sets The data set is divided into three distinct sets called training, testing and validation sets. The training set is the largest set and is used by neural network to learn patterns present in the data. The testing set is used to evaluate the generalization ability of a supposedly trained network. A final check on the performance of the trained network is made using validation set.
Steps Contd..
Neural network architecture Defines its structure including number of hidden layers, number of hidden nodes and number of output nodes. In our case, Input Nodes Hidden Nodes Output Nodes User_Age Plan_Head Plan_Name Invest_Sector Amt_invested Benefits Comp_Name Plan_Type Fund_Charges Risk Factors
Evaluation criteria The most common error function minimized in neural networks is the sum of squared errors.

Neural network training

Training a neural network to learn patterns in the data involves iteratively presenting it with examples of the correct known answers. The objective of training is to find the set of weights between the neurons that determine the global minimum of error function. This involves decision regarding the number of iteration i.e., when to stop training a neural network and the selection of learning and momentum values.
Flow of Control
Training the Data set

Developing the Training data set in a database. (Oracle 10g Enterprise Edition)
Selecting the Attributes to be trained. (User_Age, Plan_Id Plan_Fundvalue, Invcom_Id)
Training the Data set by applying different classifiers, clustering algorithms and deetermining the association rules.
Testing The Learning Algorithms

Apply Clustering Algo Apply Classification Algo Apply Attribute Selection
Test Dataset
Validation testing
Determine weights learned by tool
Apply to Test data and Validation data Compare the output with desired output
Formulate most appropriate weights
Conclusion
The field of neural network though requires a lot of research, is a highly potential upcoming trend. It is by far the most accurate form of data mining that can be implemented in any field. It is especially very beneficial in the prediction of viable and optimal financial solutions and market analysis. At present, data mining is a new and important area of research, and neural network itself is very suitable for solving the problems of data mining because its characteristics of good robustness, self-organizing adaptive, parallel processing, distributed storage and high degree of fault tolerance.
A very important feature of these networks is their adaptive nature, where learning by example replaces programming in solving problems. This feature makes such computational models very appealing in application domains where one has little or incomplete understanding of the problem to be solved but where training data is readily available. The combination of data mining method and neural network model can greatly improve the efficiency of data mining methods, and it has been widely used.
Further Work
Till now, With the help of Weka tool, we have analyzed different attributes of the tables in our database. After analyzing the output, we have compared it with the desired output and generated a report on optimum plan for a customer, thereby successfully implementing Neural Networks.
We intend to publish a research paper based on the implementation of neural networks in the field of Decision Support System in life insurance, by carrying out extensive research on implementation of neural network models.
Possible Hiccups
For some tasks, neural networks will never replace conventional methods. The weights of inputs with respect to output need to be updated with time incase of financial markets.

Dss On Life Insurance

Uploaded by

Copyright:

Available Formats

Dss On Life Insurance

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Dss On Life Insurance

Uploaded by

Copyright:

Available Formats

DSS ON LIFE INSURANCE

Revised Problem Statement

Important concepts used..

Experimenting On WEKA Tool

Development of an ANN model

Neural network training

Training the Data set

Selecting the Attributes to be trained. (User_Age, Plan_Id Plan_Fundvalue, Invcom_Id)

Testing The Learning Algorithms

Determine weights learned by tool

Formulate most appropriate weights

You might also like