Comparative Performance of CNN+LSTM, CNN+BiLSTM, and a Proposed Model

A small Excel file containing comparative performance metrics for three deep learning model architectures. The data includes mean values, standard deviations, and 95% confidence intervals. The dataset was authored by Divya Kesavulu and uploaded to figshare in April 2026.

Use Cases

• Benchmarking proposed models against established CNN+LSTM and CNN+BiLSTM baselines based on the comparative metrics.
• Analyzing the statistical significance of performance differences based on the provided confidence intervals.
• Visualizing model performance comparisons for academic papers or reports based on the tabulated results.

Strengths

• Includes statistical measures of uncertainty with standard deviations and 95% confidence intervals.
• Released under a permissive CC-BY-4.0 license for reuse and redistribution.
• File is a small 5.5 KB XLS, making it quick to download and inspect.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare, author Divya Kesavulu.

Collection Method

Likely generated from experimental results comparing neural network models.

Time Range

Publication date suggests results are from work circa 2026.

Freshness

Last updated 2026-04-21 17:49:03; freshness should be verified.

Geography

Geographic scope of the underlying data is not specified.