Triclosan (TCS) is being detected in breast milk and in infants of puerperal women. The harmful effects caused by this compound on living beings are now critical and thus it is pivotal find new tools to TCS monitoring. In the present study, an electronic tongue (e-tongue) device comprising an array of sputtered thin films based on Multi-Walled Carbon Nanotubes and titanium dioxide was developed to identify TCS concentrations, from 10−15 to 10−5 M, in both water and milk-based solutions. Impedance spectroscopy was used for device signal transducing and data was analyzed by principal component analysis (PCA). The e-tongue revealed to be able to distinguish water from milk-based matrices through the two Principal Components (PC1 and PC2), which represented 67.3% of the total variance. The PC1 values of infant formula milk powder prepared with tap water (MT) or mineral water (MMW) follows a similar exponential decay curve when plotted with the logarithm of concentration. Therefore, considering the TCS concentration range between 1015 and 10−9 M, the PC1 values are fitted by a straight line and values of −1.9 ± 0.2 and of 7.6 × 10−16 M were calculated for the sensor sensitivity and sensor resolution, respectively. Additionally, a strong correlation (R = 0.96) between MT and MMW PC1 data was found. These results have shown that the proposed device corresponds to a promisor method for the detection of TCS in milk-based solutions.