The core yarn is a type of yarn that has a filament fiber in the center with a different fiber (especially cotton) wrapped around it. They have a raising importance in the textile industry. That is why error free production and design of models that can correctly estimate the product quality parameters from fiber quality and spinning parameters are needed more and more. In the literature review, a lot of artificial neural network based estimating models for the yarn production, fabric, finishing etc. can be found. However, these prediction approach did not apply on the core yarns. The artificial neural networks can be seen to show much success in the textile. Therefore, in this study the Artificial Neural Networks (ANN) was proposed to estimate the quality control parameters of core yarns. The dataset used to feed the ANNs includes 37 types of fibre quality characteristics and spinning parameters as inputs and 11 types of core yarn performance values as outputs. There are 227 samples for each of them without any missing data. The quality characteristics of the fibre are collected from both high volume instruments (HVI) and advanced fibre information system (AFIS) machines. All the available learning algorithms for ANNs was used to get a more detailed comparison and to make sure we have a valid data different number of nodes were used for the hidden layer on the artificial neural networks. ANNs have 37 nodes of an input layer and 11 nodes of an output layer. In the architecture of ANNs, sigmoid function is used as activation function. Mean Square Error (MSE) and Mean Absolute Percentage Error (MAPE) with R squared values were used as performance indicators. The neural networks were shown to have the best MSE and MAPE values. The developed model has shown to have over 90% success rate for most of the cotton/elastane core yarn quality characteristics.