Application of text-mining techniques for extraction and analysis of paracetamol and ibuprofen marketed products' qualitative composition

Text mining (TM) applications in the field of biomedicine are gaining great interest. TM tools can facilitate formulation development by analyzing textual information from patent databases, scientific articles, summary products characteristics, etc. The aim this study was to utilize perform qualitative analysis paracetamol (PAR) and ibuprofen (IBU) formulations, terms identifying evaluating presence excipients specific active pharmaceutical ingredient (API) and/or dosage form. A total 152 were analyzed. Web-scraping used retrieve data, Python-based open-source software Orange 3.31.1 for statistical (ANOVA) obtained results. majority marketed both APIs tablets. predominant all tablet formulations povidone, starch, microcrystalline cellulose hypromellose. Povidone, stearic acid, potassium sorbate, maize starch pregelatinized occurred more frequently PAR On other hand, titanium dioxide, lactose, shellac, sucrose ammonium hydroxide IBU oral suspensions contained dispersible cellulose; liquid sorbitol; methyl propyl parahydroxybenzoate, glycerol acesulfame potassium. Specific forms, such as effervescent tablets, hard capsules, powders, solutions suspensions, well gels soft also evaluated.