Multi-channel analysis method and device based on analytic hierarchy process

Abstract

The application provides a multi-channel analysis method and a multi-channel analysis device based on an analytic hierarchy process. The method comprises the steps of: acquiring a sub-target of each channel and an index corresponding to each sub-target, building a sub-target layer paired comparison matrix table by using the sub-targets corresponding to the channels as factors, building an index layer paired comparison matrix table by using the indexes corresponding to the sub-targets of the channels as factors, computing a weight coefficient corresponding to each sub-target of each channel according to the sub-target layer paired comparison matrix table, computing a weight coefficient corresponding to each index in each sub-target of each channel according to the index layer paired comparison matrix table, acquiring a quantity value of each channel at the same index, setting an effect value of each channel at each index according to the quantity value, and weighting the weight coefficients and the effect values to obtain a final assessing score of the channel. According to the multi-channel analysis method and device based on the analytic hierarchy process, a more scientific and accurate multi-channel assessing result can be provided.

Description

technical field [0001] The present application relates to the technical field of computer analysis, in particular to a multi-channel analysis method and device based on analytic hierarchy process. Background technique [0002] In the field of marketing, it is often necessary to evaluate the value of market channels. The usual method is to quantitatively evaluate the value of the channel from a single dimension; or to evaluate the value of the channel simply and comprehensively from several dimensions. For example, carry out several promotional channel activities for a new product, and evaluate the effect of each promotional channel activity. [0003] The existing schemes include the split method and the simple comprehensive weighting method. The split method separates a certain indicator from several promotion channel activities for evaluation and comparison. For example, compare the conversion rate of new users in several promotion channel activities, and sort the results...

AI Technical Summary

Problems solved by technology

[0004] However, these two schemes have the following disadvantages: 1). Dimensionless: the scores of existing channels are calculated based on absolute or relative quantities to obtain their quantitative values, and there are differences in unit quantification for different channel quantities (ie The order of magnitude is different, there is no comparability between different data levels, and the calculation scale is inconsistent)

2). Not accurate enough: Different channels show different effects on each indicator. If you simply assign a weighted value to each indicator and then score it, or take a certain indicator for comparison, it is not accurate enough to effectively evaluate the optimal channels and promote

Images