Psychology Quality Dataset: Curated Conversations for Mental Health AI

A curated dataset for training AI models on psychology and mental health conversations. It was cleaned and filtered from multiple public HuggingFace datasets by author nayaksomkar, with a last recorded update in April 2026. The data is optimized for fine-tuning mental health chatbots.

Use Cases

• Fine-tuning mental health chatbots based on curated psychology conversations.
• Training AI assistants to generate empathetic responses based on user/assistant dialogue structures.
• Benchmarking language model performance on psychology-themed question-answering tasks.

Strengths

• Explicitly described as a curated, high-quality dataset cleaned from multiple sources.
• Optimized for a specific task: fine-tuning mental health chatbots.
• Data was converted to a standard user/assistant format, suggesting structural consistency.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and license information are unknown, limiting suitability assessment.
• Data may reflect source bias inherent to the original HuggingFace datasets used.

Provenance

Source

Aggregated and cleaned from multiple public HuggingFace datasets including Gragroo/psy_conv_v1_gpt1, Gragroo/psy_conv_v1_gpt2, Gragroo/psy_conv_v1_gpt3, psycode1/psyset, and Gragroo/psychology-question-answer_psygpt_format.

Collection Method

Curated, cleaned, and filtered from source datasets, then converted to a standard user/assistant format.

Freshness

Last updated 2026-04-15 05:48:30; freshness should be verified.