Autism Diagnosis via NLP Feature Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current natural language processing technologies lack effective methods for identifying symptoms of autism spectrum disorder through textual conversations, failing to accurately diagnose and provide real-time feedback to improve social interaction skills.
Innovation Solution
A system and method using machine learning models trained on labeled conversation data to evaluate textual conversations for autism symptoms, generating features and weights to determine the likelihood of a participant being on the autism spectrum, and providing coaching on social norms and conversational cues.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If natural language processing is applied to evaluate textual conversations for autism diagnosis, then diagnostic capability is improved, but measurement precision is insufficient
Solution Approach 1:
The conversation text is segmented into multiple linguistic features including pronoun usage, question patterns, emotional expressions, and conversational turn-taking metrics. This segmentation allows the system to analyze specific linguistic markers independently and combine them for comprehensive diagnosis, improving measurement precision while maintaining automation.
Solution Approach 2:
The system transforms raw conversation text into multiple derived parameters and features (e.g., pronoun frequency ratios, question-to-statement ratios, emotional expression density). By changing the parameter representation from raw text to structured linguistic features, the system enhances diagnostic precision while preserving automated evaluation.
2Measurement precision
If machine learning models are trained on labeled conversation data, then diagnostic accuracy is improved, but device complexity increases
Solution Approach 1:
The system performs preliminary actions by pre-training machine learning models on extensive labeled conversation datasets before deployment. This preliminary training phase extracts and stores learned patterns and weights, so that during actual diagnosis, the system only needs to apply pre-computed models to new conversations, reducing real-time complexity while maintaining high diagnostic accuracy.
Solution Approach 2:
The system uses copying by training on labeled examples and creating a replicated model that can be deployed independently. The learned patterns from training data are copied into the model structure, allowing the system to achieve high diagnostic accuracy through the trained model without requiring the original training infrastructure during deployment, thus managing complexity.
3Productivity
If real-time feedback is provided on conversational patterns, then productivity of social skill improvement is improved, but loss of time for processing increases
Solution Approach 1:
The system applies partial action by providing real-time feedback on specific problematic conversational patterns rather than analyzing every aspect of the conversation. It focuses on key indicators such as pronoun usage errors, missing emotional expressions, or inappropriate question patterns, delivering targeted feedback that improves social skills efficiently without requiring complete conversation re-analysis, thus reducing processing time while maintaining productivity.
Data Source
AI summary
Embodiments herein include a natural language computing system that provides a diagnosis for a participant in the conversation which indicates the likelihood that the participant exhibited a symptom of autism. To provide the diagnosis, the computing system includes a diagnosis system that performs a training process to generate a machine learning model which is then used to evaluate a textual representation of the conversation. For example, the diagnosis system may receive one or more examples of baseline conversations that exhibit symptoms of autisms and those that do not. The diagnosis system may annotate and the baseline conversations and identify features that are used to identify the symptoms of autism. The system generates a machine learning model that weights the features according to whether the identified features are, or are not, an indicator of autism.