Table 4.

A comparison of the existing studies present in the literature that are based on text dataset classification

Author	Dataset	Algorithm/Model	Accuracy (in %)
Wang et al. [26]	Mayo Clinic smoking	CNN	92
Wang et al. [26]	Proximal femur (hip) fracture	CNN	97
Wang et al. [26]	i2b2 2006 smoking	Rule-based NLP	89
Nguyen et al. [28]	Language Resources and Evaluation Conference (LREC)	Back off	66.39
Hughes et al. [27]	Merck Manual	CNN + Word2vec	68
Proposed method	Hospital records	KMP	97.4

Wang et al. stated that CNN gave the best accuracy for text classification. Deep learning models are seen to be commonly used for this purpose. It is observed that the proposed method gave the highest accuracy