In this work, response surface methodology and adaptive neuro-fuzzy inference system approaches were used to predict and model effect of extraction conditions of pectin from medlar fruit (Mespilus germanica L.). The pectin extracted at optimized conditions (89 °C, 4.83 h and 4.2 pH) could be classified as high methoxyl pectin. Sugar composition analysis showed that pectin was mainly composed of D-galacturonic acid, L-arabinose, L-rhamnose, D-galactose and D-glucose. Fourier Transform Infrared Spectroscopy, RAMAN and nuclear magnetic resonance spectra confirmed molecular structure, revealing presence of D-galacturonic acid backbone. X-ray diffraction patterns revealed an amorphous structure. Differential scanning calorimetry showed endothermic (123 °C) and exothermic peaks (192 °C). Thermogravimetric analysis revealed three decomposition regions, 50-225 °C, 225-400 °C and 400-600 °C. Steady and dynamic shear analyses revealed that pectin had a pseudo-plastic behavior with storage (G') and loss (G″) modulus increasing with increment in frequency, indicating viscoelastic structure more predominantly elastic than viscous.