Crop rotation is a fundamental sustainable practice in agriculture for improving soil fertility, reducing pests and diseases, and optimizing the crop yields. Traditional methods in crop rotation optimization strongly rely on expert knowledge, examining historical data which may not be fully cover the complexities of modern agricultural land. Natural language processing techniques such as text mining, sentiment analysis, language translation, and topic modeling can be used for agricultural text data analysis, decision support systems, farmer knowledge integration, predictive modeling, sustainability, and environmental impact. This paper analyzes the various NLP driven models that uses textual data from various sources for crop rotation strategy. This paper also examines the challenges and limitations of implementing NLP in agriculture such as quality of data, computational requirement and need for domain specific knowledge. By reviewing the recent advancements this paper provides an insight into future direction and potential of NLP in agricultural crop rotation optimization.