Clustering and Regression to Impute Missing Values of Robot Performance

A hybrid machine learning system to impute and classify a component-based robot

“…However, the previous work on this data proposed the elimination of MV [3] rather than their imputation. The problem addressed in this paper have already been dealt with in less depth [1], where the work focused only on cluster regression but without performing classification tests on the data resulting from the imputation. Likewise, the study was conducted only with the k-means clustering technique and in present work it has been extended with additional techniques.…”

“…The Figure 1 shows the steps of each of the stages of this hybrid System. For the selection of clustering techniques based on [1], only k-means and clustering techniques were used, although as detailed in the future work, the inclusion of hierarchical and density-based methods could be interesting. Secondly, for imputation techniques in [2] was concluded that, in general, the regression techniques that worked best were those used in the present investigation.…”

“…The robotic systems cover different disciplines like cinematic, mechatronics, electronics, and Artificial Intelligence (AI). Regarding the latter, the European Commission recently reported that 42% of enterprises use at least one technology related to this field 1 . This percentage will increase in the coming years, showing a clear need for companies to adopt these technologies, which are already consolidated in the market.…”

“…In order to address this problem, a novel Hybrid Artificial Intelligence System (HAIS) is proposed. 1 European Commission. European enterprise survey on the use of technologies based on artificial intelligence (July 2020).…”