Machine learning is an area that helps to predict outcomes more accurately. It was utilized in different domains such as banking, healthcare, education, etc. Among all the domains, machine learning was largely utilized in the healthcare sector for predicting and diagnosing the disease in advance for saving millions of lives. ML has different kinds of algorithms which help to make the prediction process effective. This chapter focussed on explaining different machine learning algorithms for making better predictions in pregnancy complications in the healthcare domain. In general, there are different complications that women encountered during their pregnancy periods such as High BP, preeclampsia, anemia, etc. This work specifically aims to describe the preeclampsia complication during pregnancy. In machine learning, various kinds of regression algorithms are compared and analyzed. It also focused on which predictive technique would be more efficient for predicting the condition of preeclampsia in advance to save lives of pregnant women and also take necessary precautions.
Top1. Introduction
Large data sets are a rich resource from which data mining may be used to uncover possible new and useful knowledge. Using data mining through Visualization, machine learning, and other data manipulation and information extraction techniques are becoming a more and more prominent field that is used to get an understanding of the linkages and patterns buried in the data (Bellary et al., 2010). A large-scale information system today incorporates separate databases or information systems. The volume of data is growing, which makes it harder to get usable data for decision support (Nti et al., 2022). Traditional manual data analysis is no longer sufficient, and to analyze and gather the necessary data from the vast amount of information, the need of technologies created in the discipline of cognitive analysis of information. There arise several techniques of machine learning and data mining with the combination of various algorithms to process those data (Majumdar et al., 2016).
As the quality of the system continually improves, Artificial intelligence (AI) and computer science have spawned a subfield known as machine learning (ML), which employs data and algorithms to simulate the way humans learn (Jena et al., 2021). ML is defined as one of the prominent areas which utilize various kinds of algorithms and techniques for carrying out predictions and classifications as well as to discover the primary factors in data mining areas (Ul Hassan et al., 2018). In general, the machine learning algorithms are categorized as supervised, unsupervised, semi-supervised, and reinforcement learning which has been depicted through Figure 1. The choices taken in response to these insights should, ideally, have an impact on key growth metrics in applications and companies. In a nutshell, ML could be defined as predicting the outcomes from a huge amount of testing data without being explicitly programmed (Li, 2020). Data scientists will be sought after more and more as big data develops and gets better. They will be required to contribute to the process of determining the most appropriate business queries and the information required to solve those questions (Jain & D. V., 2021).
Data mining (DM) along with machine learning holds a key part in many sectors and domains of the modern world. In a nutshell, Knowledge discovery in data (KDD), describes the action of removing trends, patterns as well as other significant information from large datasets, could be used to define data mining (DM) (Tawfik et al., 2022). With the advent of Data warehousing and its utilization in various firms and organizations, the application of data mining has increased rapidly for making better decisions. Insightful data analysis produced by data mining has enhanced corporate decision-making (Takeuchi et al., 2006). The two primary goals of the data mining techniques used to support these studies are to either characterize the target dataset or forecast outcomes using machine learning algorithms (Chauhan & Jangade, 2016; Kaur & Dhariwal, 2021).
Machine learning has been utilized in various arenas such as banking, healthcare, the education sector, e-commerce, recommender systems, business, etc., Meanwhile, the main utilization of ML in recent days would be focused on the healthcare sector (Renuka Devi et al., 2022).