In this study, the application of chemometrics-assisted UV spectrophotometry was presented for the simultaneous determination of Risperidone (RIS) and Haloperidol (HP) in binary mixtures and biological fluid. The basis of the spectrophotometric procedure were the least squares support vector machine (LS-SVM), fuzzy inference system (FIS), and adaptive Neuro-fuzzy inference system (ANFIS) methods. A leave-one-out (LOO) method was used to find the best value of parameters, including regularization parameter (γ) and width of the function (σ2) related to the radial basis function (RBF) kernel in LS-SVM procedure. The optimum γ and σ were obtained 11600, 17000 and 7700, 2000 for RIS and HP, respectively. Mean percent recoveries of RIS and HP in synthetic mixtures were obtained 101.72 and 99.77, respectively. The root mean square error (RMSE) related to the FIS and ANFIS models were found 0.878, 2.124 and 0.285, 0.206 for RIS and HP, respectively. Analysis of variance (ANOVA) test indicated that there were no significant differences between the proposed methods and high-performance liquid chromatography (HPLC) as a reference technique. The obtained satisfactory results demonstrate that the use of the LS-SVM and ANFIS models coupled with UV–vis spectrophotometry are a promising idea for industrial quality control and analysis of pharmaceutical formulations.

在这项研究中，提出了化学计量学辅助紫外分光光度法在同时测定二元混合物和生物液体中利培酮（RIS）和氟哌啶醇（HP）中的应用。分光光度法的基础是最小二乘支持向量机（LS-SVM），模糊推理系统（FIS）和自适应神经模糊推理系统（ANFIS）方法。留一法（LOO）方法用于查找参数的最佳值，包括LS-SVM过程中的正则化参数（γ）和与径向基函数（RBF）内核相关的函数宽度（σ2）。RIS和HP的最佳γ和σ分别为11600、17000和7700、2000。合成混合物中RIS和HP的平均回收率分别为101.72和99.77。对于RIS和HP，与FIS和ANFIS模型相关的均方根误差（RMSE）分别为0.878、2.124和0.285、0.206。方差分析（ANOVA）测试表明，所提出的方法与作为参考技术的高效液相色谱（HPLC）之间没有显着差异。获得的令人满意的结果表明，将LS-SVM和ANFIS模型与UV-vis分光光度法结合使用是工业质量控制和药物制剂分析的有前途的想法。