Market segmentation is an extremely valuable and applicable concept in marketing. The small-to-medium enterprises (SMEs) find segmentation strategy especially beneficial, because they are closer to their buyers. A great deal of tools are around for doing market segmentations. But, the majority of tools aren’t created specifically with the needs and constraints of the SMEs in mind. Furthermore, the majority of them are designed to work with massive amount of data and it is doubtful if the SMEs can use them. The SMEs don’t have the time and resources to formulate the best model for segmentation, and they’ve to trade off between the information quality and the available resources. Survey data are their most critical information source as they are unlikely to have large customer database. Thus the SMEs are looking for a ‘quick-fix’ solution that depends on ad hoc collection of data through survey for developing their market segmentation models. It is helpful if they know which classifiers work best for such purpose. From the literature review, it is obvious that there is no standard agreement about the best classification techniques to recommend for all occasion, as whose performance is impacted by the data types and the application conditions. As classifiers perform best with various types of data, the real research goal for many comparison studies is not to look for the best method in general. Instead the researchers should find out which method works best for some specified data sets (along with their assumptions about the noise, outliers, etc.). In this research, we concentrate on comparing the performances of numerous statistical tools and artificial neural networks (ANNs). It is our main objective to find out which of them are more ideal for solving market segmentation problem for the SMEs with survey data…..
There are various studies comparing different classifiers using continuous multivariate normal data. They are not so helpful for finding out the best classifiers from survey data that comprise of both continuous and discrete variables. For example, in our real data set using the data from a survey for the package tour industry, we discover that the segmentation variables consist of mixed data type with a multivariate normal, a Poisson and a point binomial distribution. Moreover, most previous data for comparing classifiers do not contain noises or outliers. Such comparisons thus cannot be applied to real-life situations that typically take care of data contaminated with noise and outliers. Noises are common as they represent measurement errors (e.g. the respondents have recalled the frequency of purchase wrongly). The outliers arise occasionally and they’re simply observations that don’t belong to any classes. To obtain a good market segmentation result we require a classifier that is strong to both noises and outliers. To figure out the best classifiers, it is helpful to compare the classifiers with both real data and simulated data. We use real-life survey data to provide information for the overall data structure and contextual meaning for the identified segments that is beneficial for the evaluation of segments in a practical sense. Real-life data are useful in showing how efficient these segmentation tools are in offering the useful information for the development of market segments. On the flip side, data from the Monte Carlo simulation are utilized to compare the efficiency of true class recovery by various classifiers. Simulated data with parameters determined from real-life data are helpful for selecting the best classifiers as their true classes are known beforehand. In our research, comparisons were made using simulated data with carefully chosen parameters for various experimental designs. Statistical significance can be linked to the tests and the outcomes thus can be generalized to similar situations…..
Source: City University of Hong Kong