Community detection in attributed networks aims to recover clusters in which the within-community nodes are as interconnected and as homogeneous as possible, while the between-communities nodes are as disconnected and as heterogeneous as possible. The current research proposes a straightforward data-driven model with an integrated regularization term to recover communities. For further improvement of the quality of detected communities we also propose a softmax-scaled-dot-product to transform the data spaces into more cluster-friendly data spaces. We adopt the gradient descent optimization strategy to optimize our proposed clustering objective function. We compare the performance of the proposed method using both real-world and synthetic data sets with three state-of-art algorithms. Our results showed that the proposed method obtains promising result.