Evaluations

The Subtxt Narrative Framework is designed to support narrative analysis by identifying the core sources of conflict within storytelling. With this, writers and developers can evaluate key elements for thematic coherence, enabling more compelling, cohesive stories. The Evaluations endpoint makes this framework accessible for integration into applications, allowing users to objectively measure and refine narrative components according to thematic conflict sources.

Using the Evaluations Endpoint

To analyze a storytelling element, you can submit a POST request to the Evaluations endpoint. This request requires specifying a conflict method and a block of storytelling text, which Subtxt will evaluate to determine if the narrative effectively illustrates the thematic conflict. A higher score indicates that the storytelling is a strong match for the chosen method of conflict.

API Endpoint

URL: https://api.subtxt.app/api/v1/evaluate/storytelling

Parameters

The evaluate/storytelling endpoint requires two parameters:

Request Example

curl -X POST https://api.subtxt.app/api/v1/evaluate/storytelling \
-H "Authorization: Bearer your-token-here" \
-H "Content-Type: application/json" \
-d '{
        "method_of_conflict": "Disbelief",
        "storytelling": "Josephine tells Margaret that she is totally skeptical of anyone she has never met before, and would prefer to make her own decisions."
    }'

Sample Responses and Interpretation

Sample 1: A Strong Match

In this example, the request includes Control as the method_of_conflict and storytelling text that reflects a clear exercise of control in a dramatic context.

{
    "method_of_conflict": "Control",
    "storytelling": "directing the group towards certain death"
}

Response:

{
    "approved": true,
    "evaluation": "Good illustration",
    "score": 0.863622427
}

A high score of 0.863622427 indicates that the narrative effectively captures the theme of Control, reinforcing the thematic intent.

Sample 2: An Average Match

Here, the method_of_conflict is set to Disbelief, with storytelling that conveys some skepticism but lacks depth in illustrating the conflict.

{
    "method_of_conflict": "Disbelief",
    "storytelling": "Josephine tells Margaret that she is totally skeptical of anyone she has never met before."
}

Response:

{
    "approved": true,
    "evaluation": "Passable",
    "score": 0.85238117
}

This score indicates the storytelling is adequate but may benefit from deeper exploration of the Disbelief theme to better align with the conflict type.

Sample 3: A Weak Match

In this example, Actuality is chosen as the method_of_conflict, with storytelling that does not clearly convey the conflict type.

{
    "method_of_conflict": "Actuality",
    "storytelling": "Josephine freaks out and runs around the library like a maniac."
}

Response:

{
    "approved": false,
    "evaluation": "Complete denial",
    "score": 0.777258813
}

Here, the lower score (0.777258813) and the evaluation of "Complete denial" suggest the storytelling does not adequately illustrate the Actuality conflict, indicating a need for refinement.

Summary

The Evaluations endpoint offers an objective assessment of thematic conflict within storytelling. By using the method_of_conflict and storytelling parameters, you can integrate powerful thematic evaluations into your narrative tools and workflows. This endpoint not only helps assess storytelling elements but also serves as a guide to enhance the depth and effectiveness of thematic representation in narratives.