I have a dataset which consists of categorical data like job roles(eg. 'Engineering Coordinator': 235122836, 'Production Enablement Engineer': 737188091, 'Engineering Manager: Software': 946399313, 'VP of Operations': 428182487, 'Analytics Manager': 1018924788, 'Director of Sales': 421483035 )
I have performed encoding on this data and used it for outlier detection with Local Outlier Factor but these encodings are causing bias in my output as when I change these encoded values, the output changes.
Can someone please suggest a good way to encode these categorical features