The widespread use of technology applications, artificial intelligence and aggressive trends towards the knowledge economy in general have led to the creation of large-scale data repositories, and these large data repositories will make a significant contribution to future decision-making, as appropriate knowledge detection mechanisms are applied to extract hidden information from such data, which can be invaluable for finding useful hidden patterns and trends and producing new observations on data.
It is self-evident that the vast amount of data available in the information industry will remain useless until it is converted into useful information (which was later called knowledge mining) and this technique confirms that the overall goal is to extract patterns and knowledge from large amounts of data, not mining from the data itself.
The first roots of the term cognitive mining were in the 1960s and they were used by economist Michael Lovell in an article published in the 1983 Economic Studies Review.
The term was more clearly demonstrated in 1990 in the database community, with positive alternatives, and in the academic community; the main research forums in cognitive mining began in 1995 when the first International Conference on Data Mining and Knowledge Discovery (CD-95) was established in Montreal, and a year later the Conference was launched by the Journal of Data Mining and Knowledge Discovery (Cognitive Mining). Later, it became the first magazine in this field.
Following the development of cognitive mining technology, it has been used to detect hidden information, decision-making and predict future trends in financial markets, and the competitive advantages of this technology in business and finance include increased revenues, low costs, market improvement and response, and awareness mode.
The term cognitive mining involves basic steps that are necessary in any cognitive mining process to achieve knowledge discovery, the first of which is data cleaning, where noise and inconsistent data are removed; this is followed by data integration and multiple data sources that are combined as the data selection phase comes in, by which data related to the analysis task is recovered from the database, then data is converted and in this step, it is converted or integrated into appropriate forms of mining through the implementation of summary or assembly processes, and then the extraction of data, where intelligent methods are applied in order to extract data patterns, followed by the step of evaluating data patterns, then the step of displaying knowledge in which knowledge is represented and, once all these processes are completed, we will be able to use this information in many applications, such as fraud detection, market analysis, production control, science exploration, as well as any application of a supported system associated with computer uses, including artificial intelligence, machine learning and commercial intelligence. In other words, we can say that data mining is about the procedure of data mining.
The task of extracting actual data is either the semi-automatic or automatic analysis of large amounts of data to extract previously unknown patterns; it is therefore interesting as sets of extraordinary data, records, and reliability. This usually involves the use of database techniques such as spatial indicators. These patterns can then be seen as a kind of input data summary. They can be used for further analysis or, for example, in machine learning and predictive analysis.
Cognitive mining is a multidisciplinary technique in which the focus is on useful knowledge extraction methodologies from data, the continued rapid growth of data on the Internet and the widespread use of databases, which have created a tremendous need for methodologies. The biggest challenge is first extracting knowledge from data and then researching statistics and databases, pattern recognition, machine learning, data visualization, optimization, and high-performance computing, to deliver advanced business intelligence solutions and web discovery.
Finally, it can be said that the tremendous growth of online data has made cognitive mining-based solutions increasingly important to service and consulting companies, and accordingly, the systematic development of business intelligence, as well as the IT system and business process controls, which have become a focal point for statistics and research by major knowledge mining companies, and monitoring data collected over time is used to make processes both efficient, effective, predictable and profitable.
Difficult aspects of cognitive mining include dealing with large time-based data with a variety of characteristics, producing accurate and practical methods of forecasting, and developing business-related decision-making analyses. And to talk the rest.
Author : Manahel Thabet
Published February 19, 2018
Al Bayan Newspaper
AR 
EN