Dark Mode
Pharmaceutical Patient Feedback
Product Reviews & Feedback
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This dataset offers a focused collection of patient reviews on specific drugs, coupled with details about their associated medical conditions. It is designed to facilitate the analysis of drug experiences, pinpointing top-rated medications and uncovering insights into benefits and side effects. The data supports sentiment analysis of drug reviews, enabling researchers to explore model transferability across various health conditions and different data sources within the pharmaceutical domain.
Columns
- reviewID: A unique numerical identifier assigned to each individual patient review.
- urlDrugName: The textual name of the drug under review, indicating the specific medication patients have provided feedback on.
- condition: The medical ailment or condition for which the drug was prescribed or used.
- benefitsReview: Textual descriptions from patients detailing the positive effects or advantages they experienced while taking the medication.
- sideEffectsReview: Textual descriptions from patients outlining any adverse reactions or negative effects encountered due to the drug.
- commentsReview: General textual feedback and overall observations provided by patients regarding their experience with the medication.
- rating: A numerical value, on a scale of 1 to 10 stars, representing the patient's overall satisfaction or perceived effectiveness of the drug.
- sideEffects: A categorical variable indicating the reported severity or impact of side effects, categorised into five distinct levels.
- effectiveness: A categorical variable reflecting the patient's perceived effectiveness of the drug in treating their condition, also categorised into five distinct levels.
Distribution
This is a multivariate dataset that incorporates both textual and numerical data types. It comprises 4143 individual instances (records) and is characterised by 8 distinct features. The data is structured into two tab-separated-values (.CSV) files: a training set which constitutes 75% of the data and a test set representing the remaining 25%. The files include
drugLibTest_raw.csv (795.62 kB) and drugLibTrain_raw.csv (2.29 MB). A key characteristic is the absence of any missing values within the dataset.Usage
This dataset is ideally suited for a variety of analytical and machine learning applications, including:
- Sentiment analysis: Understanding patient sentiment towards various drugs and their effects.
- Classification tasks: Predicting drug effectiveness, side effect severity, or patient satisfaction categories.
- Regression analysis: Modelling factors influencing patient drug ratings.
- Clustering: Grouping similar drugs or patient experiences based on review content and ratings.
- Identifying top-rated medications: Discovering drugs highly valued by patients for specific conditions.
- Research on model transferability: Investigating how analytical models perform across different medical conditions or data sources.
Coverage
The dataset's scope is primarily focused on patient feedback regarding drug efficacy and side effects for various medical conditions. The original data was obtained from online pharmaceutical review sites, Druglib.com and Drugs.com. While it captures diverse drug experiences, specific demographic information (such as age, gender, or geographic location) of the patients providing reviews is not included. The time range of the collected data is not explicitly stated.
License
Attribution 4.0 International (CC BY 4.0)
Who Can Use It
This dataset is particularly valuable for:
- Researchers and academics: Engaged in studies within health informatics, pharmacology, data science, and natural language processing.
- Data scientists and machine learning engineers: Developing predictive models for drug outcomes or sentiment analysis tools.
- Public health analysts: Investigating drug safety and patient reported experiences.
- Pharmacists and healthcare professionals: Seeking insights into patient perceptions of medications.
- Any individual or organisation conducting non-commercial research into patient drug experiences.
Dataset Name Suggestions
- Patient Drug Review Sentiment
- Medication Efficacy & Side Effects
- Healthcare Drug Experience Data
- Pharmaceutical Patient Feedback
- Drug Review Analysis Dataset
Attributes
Original Data Source: Pharmaceutical Patient Feedback
QUALITY