Weightings List

Displays a set of TF-IDF weightings, along with a header containing totals.

Remarks

The Weightings List control provides a detailed, tabular view of term frequency-inverse document frequency (TF-IDF) weightings used in Lexical classification and Field Classes. It is designed to help users review, interpret, and compare the relative importance of features (such as words or tokens) across a set of training data or classification results.

Purpose and Functionality

• Weightings Table:
  Presents a sortable, filterable list of features and their associated TF-IDF scores, allowing users to analyze which features contribute most to classification or scoring.
• Header with Totals:
  Displays summary information above the table, including total occurrences, total classes, and the number of training items for the current class or model.
• Feature Vector Integration:
  Highlights or annotates features present in the current feature vector, supporting rapid identification of influential terms.
• Dynamic Sorting:
  Entries are sorted by score (descending) by default, making it easy to focus on the most significant features.
• Contextual Display:
  The control adapts to the presence or absence of training data, and can optionally show class frequency (CF) information when available.

UI Components

• Weightings Table:
  • Tabular list of features (e.g., words, tokens) and their TF-IDF scores.
  • Columns may include feature name, score, and additional metadata.
  • Supports row selection, sorting, and filtering.
• Header Section:
  • Displays totals for occurrences, classes, and training items.
  • May include contextual captions or class-specific information.

Example

The following diagram illustrates the layout of the Weightings List control:

┌────────────────────────────────── Weightings List ─────────────────────────────────────┐
│ Total Occurrences: 1,234   Total Classes: 5   Class Training Items: 200                │
│----------------------------------------------------------------------------------------│
│ {Feature List}                                                                         │
│ ┌────────────────────────────────────────────────────────────────────────────────────┐ │
│ │ Feature   | Weighting | Count | CWF | CTC | CF       | TF      | IDF     | ...     │ │
│ │-----------|-----------|-------|-----|-----|----------|---------|---------|---------│ │
│ │ Invoice   | 0.980000  |  120  |  3  |  80 | 0.800000 | 0.500000| 1.200000| ...     │ │
│ │ Amount    | 0.850000  |   95  |  2  |  60 | 0.600000 | 0.400000| 1.100000| ...     │ │
│ │ Due       | 0.750000  |   80  |  2  |  55 | 0.550000 | 0.350000| 1.050000| ...     │ │
│ │ ...       | ...       |  ...  | ... | ... | ...      | ...     | ...     | ...     │ │
│ └────────────────────────────────────────────────────────────────────────────────────┘ │
└────────────────────────────────────────────────────────────────────────────────────────┘

• The header displays the caption and summary totals.
• The table lists features and their TF-IDF scores, sorted by importance.

Usage Tips

• Use the table to identify which features are most influential in the current classification or scoring model.
• Review the header for context about the size and scope of the training data.
• Sort or filter the table to focus on specific features or score ranges.
• Use this control to validate feature selection, interpret model results, or troubleshoot classification performance.

The Weightings List control streamlines the process of reviewing and interpreting feature importance in Grooper, providing clear, actionable insights for solution designers, data scientists, and reviewers.

Child Controls

┌──────────── Weightings Table ─────────────┐
│ Feature   | Score  | Count | CF | ...     │
│-----------|--------|-------|----|---------│
│ invoice   | 0.98   |  120  | 80 | ...     │
│ amount    | 0.87   |   95  | 60 | ...     │
│ due       | 0.75   |   80  | 55 | ...     │
│ ...       | ...    |  ...  | .. | ...     │
└───────────────────────────────────────────┘

Used By

Content Type - WeightingsField Class - Weightings