In the contemporary context, due to the high volume of unlabelled data in fields such as medicine, agriculture, chemistry, and many more; unsupervised machine learning models have been topics of interest and of investment. One of the computationally inexpensive and fast models investigated in this paper will be the fuzzy form of K-Means known as Fuzzy C-Means (FCM). Since FCM like K-Means requires the cluster number beforehand it is also vital that the cluster validity indices be fuzzy. In this paper, the evolutionary steps of FCM will be compared by evaluating the models suggested to overcome the pitfalls of the FCM algorithm. As there are many other algorithms created for this purpose, the algorithms analyzed in this article will be Possibilistic C-Means (PCM) and Unsupervised Fuzzy Possibilistic C-Means (UFPC). The comparison of these models is crucial since the new parameters introduced affect the cluster number chosen as seen in the internal validity indices. For applying the algorithms 4 benchmark datasets will be studied in R that belong to fields from biology, chemistry, and demography. The researcher expects that the UFPC algorithm will surpass the others since, the algorithm uses parameters from both FCM and PCM, however, as real-life datasets are rather complex, it is significant that the analysis be compared to benchmark datasets as proposed in this article. The performance will be evaluated on 12 fuzzy clustering validity indices and 3 internal validity indices that being silhouette, gap, and WSS. Custom R libraries will be used to ease the process of applying the algorithms and validity indices.