In UiPath Communications Mining, over-reliance on the Search feature during the training process can lead to an increase in model bias. This happens because using search-based filtering to identify and label examples might not represent the full diversity of the data. The model could be trained on a skewed subset of the data, causing it to favor certain patterns or keywords, and thus biasing the model towards specific types of data rather than learning to generalize effectively across all data.
Limiting the use of search ensures that the training process considers a broader and more representative sample of the data, which reduces the risk of introducing bias into the model and helps it generalize better to new, unseen communications.
For more details, refer to:
UiPath Communications Mining Documentation: Model Training and Avoiding Bias

Reference: UiPath Document Understanding Workflow

In UiPath's Taxonomy Manager, field rules are used to optimize extraction results and perform automatic validation of extracted data. These rules can be applied to specific fields within the document understanding process to ensure that extracted data meets certain predefined conditions or constraints, such as formats, patterns, or value ranges. This allows for higher accuracy in the extraction process and reduces errors, as invalid data can be flagged or corrected based on these rules.
Field rules help automate the validation of data by setting criteria that extracted data must meet, thus enhancing the extraction quality and ensuring that only valid and structured data is processed further in the automation workflow.
For more details, refer to:
UiPath Taxonomy Manager Documentation: Field Rules in UiPath
UiPath Document Understanding Framework: Taxonomy Manager Overview

Explanation:
The correct execution order of the Document Understanding template stages is:
Taxonomy
Digitize
Classify
Extract
Extraction Validation
Export
Comprehensive and Detailed Explanation: The Document Understanding template stages are based on a document processing flowchart that follows these steps:
First, you need to define the Taxonomy of the document types and fields that you want to process and extract information from. This is done using the Taxonomy Manager in UiPath Studio1.
Next, you need to Digitize the input documents, which can be in various formats such as PDF, image, or text. This is done using the Digitize Document activity, which converts the documents into a machine- readable format and performs OCR if needed2.
Then, you need to Classify the digitized documents into the predefined document types in your taxonomy. This is done using the Classify Document Scope activity, which can use various classifiers such as Keyword Based Classifier, Machine Learning Classifier, or Intelligent Form Extractor3.
After that, you need to Extract the relevant information from the classified documents based on the fields in your taxonomy. This is done using the Data Extraction Scope activity, which can use various extractors such as Regex Based Extractor, Machine Learning Extractor, or Form Extractor.
Next, you need to perform Extraction Validation to review and correct the extracted information, either manually or automatically. This is done using the Present Validation Station activity, which can use either the Validation Station or the Action Center for human-in-the-loop validation.
Finally, you need to Export the validated information to the desired output location, such as a file, a database, or a queue. This is done using the Export Extraction Results activity, which can use various exporters such as Excel Exporter, CSV Exporter, or Queue Item Exporter.
References:
UiPath Studio - Taxonomy Manager
UiPath Activities - Digitize Document
UiPath Activities - Classify Document Scope

The Shuffle training mode in the "Explore" phase of UiPath Communications Mining helps to introduce randomness into the training data. This prevents labeling bias by ensuring that the model does not overly focus on certain labels or examples, leading to a more balanced and generalized model. By shuffling the data, the model is exposed to a diverse range of examples, which enhances its ability to make accurate predictions across different labels and datasets.
(Source: UiPath Communications Mining documentation)