Planar chromatography and related techniques [micro-planar chromatography, micro-TLC, or paper-based microfluidic devices (μPADs)] present several advantages in analytical applications, such as simplicity, low cost of analysis, and the ability to work with raw complex samples without the involvement of time-consuming prepurification steps. By using commonly applied planar chromatographic systems and μPADs devices, stationary phases (silica and cellulose based), different solvent mixtures (methanol-water and dichloromethane-methanol), and proportions varying from 0 to 100% (v/v), micro-TLC migration profiles of several dyes described in terms of characteristic of chromatographic parameters (retardation factor, peak base width, and asymmetry factor) were investigated. Combining these results with some quantum mechanics calculated properties for each solute (dipole moment, polarizability), and by using the data mining approach, we modeled this overall chromatographic behavior in order to describe experimental data. With this approach, we were able to predict with reasonable confidence some chromatographic properties. This effort its crucial in order to (1) optimize solute elution, (2) increase mixture resolution, and (3) identify some molecular properties of analytes for designing simple micro-TLC. It is hoped that the presented nonhypothesis-driven data-mining approach can be helpful for understanding the chromatographic behavior of dyes on silica and cellulose adsorbents using the simplest mobile phases. This should be helpful for further designing the micro-TLC separation systems or μPADs quantification devices based on cellulose and related biopolymers and considering dye compounds as analytes for separation and sensing molecules.