Digital Taxonomy Information Center Support Center

Contact Us

Autocoding Mode

Introduction
The Admin section within Codeit offers an "Autocoding Mode" setting on each Task.
This setting can be used to regulate whether the Codeit AI can automatically apply codes generated by the Machine Learning based on the confidence of the suggestions available for each item. This page explains what the options are for this setting and how each behaves.
Note that this setting only applies to the Machine Learning layer of the Codeit AI.

1. Complete Coding
The aim of the Complete Coding mode is ensure that the AI can only apply suggestions from the machine learning if all of the suggestions are of a suitable quality. To understand this, consider the following example:

Suppose that the Codeit Machine Learning Layer is presented with the verbatim:
"They give good service, but it was expensive and the staff were rude"

Suppose also that the Machine Learning generates the following suggestions, in response:

SuggestionConfidence
Code 3: Good Service98%
Code 7: Poor Price/Value / Too Expensive91%
Code 17: Queuing Issues15%


Clearly, the confidence level of the last suggestion is quite low.
In Complete Coding mode, the confidence of all of the suggestions must be above, or equal to, the "Autocoding Threshold" for any Machine Learning suggestions to be applied.
The rationale for this is that suggestions with a low confidence suggest that there is something in the verbatim that the Machine Learning has missed, or not got right.
In this example, if we only apply the first two codes, then we are likely to have a gap in our coding - in this sense the codes applied are not considered "complete".
Using Complete Coding, in this case, would result in none of these suggestions being applied.
So, you would use Complete Coding mode if you want to be as sure as possible that applying the Machine Learning suggestions will result in completely coded items with no gaps of this sort.
Note, that this sets quite a high bar for the AI and will result in higher "completeness" of autocoding at the expense of a lower number of items autocoded overall.

2. Partial Coding
The Partial Coding mode offers an alternative approach - we allow the Machine Learning Layer to autocode any suggestions that meet the minimum "Autocoding Threshold".
This increases the possibility of gaps in the coding, but hugely increases the volume of codes that the AI is able to apply from the Machine Learning.

Powered by HelpSite