The glass identifier bot helps to investigate broken glass sources – for example, for criminology purposes – by analysing the glass refractive index and the composition. The bot has been trained with 214 samples from [1] and was cross-validated by using the same dataset.

The following features are required when using the bot:

• Refractive index
• Sodium (weight percent in corresponding oxide)
• Magnesium
• Aluminum
• Silicon
• Potassium
• Calcium
• Barium
• Iron

The glass will be classified as one of the following labels:

• [building_windows1]: Building windows float processed
• [building_windows2]: building windows non float processed
• [containers]
• [vehicle_windows]
• [tableware]
• [headlamps]

Copy and paste the code to embed this bot on your site:

<iframe style="max-width: 100% !important; border: none; height:180px" width="560" src="https://hailp.com/api/main/loadwidget/483/parent/csvbulk"></iframe>

TRY ME, FILL IN THE FIELDS BELOW

Credit: Data from B. German, Central Research Establishment, Home Office Forensic Science Service – Aldermaston, Reading, Berkshire RG7 4PN