PsyProxy
Datasets·Medicine·02__health_reviews__druglib__benefits_rating

Druglib drug-overall-rating reviews

Druglib Drug Reviews is a public corpus of patient-written reviews of prescription medications, each rating a drug’s benefits and overall effectiveness on 1–10 ordinal scales. The texts predominantly discuss a range of personal experiences related to the effects of various medications, highlighting both positive and negative outcomes. Common themes include improvement in mood , relief from physical pain , and management of chronic conditions such as anxiety, insomnia, and acne. Many reviews express satisfaction with treatments that resulted in reduced symptoms like cramps , hypertension , and menstrual issues , while others detail adverse effects, including emotional instability and relationship problems . Overall, the narratives reflect a complex interplay of benefits and drawbacks, illustrating the varied impact of medications on individuals' quality of life. [Summary on 50 random texts by ChatGPT 4o Mini].

Distribution of 1–10 benefits rating
1
2
3
4
5
6
7
8
9
10
401 at floor968 at ceiling
4,120
items
824
holdout n
1–10 benefits rating
target
Continuous
kind
26
systems compared
Criterion validity

Reported holdout systems from the verified card

Continuous regression uses as the task-primary metric. Secondary columns keep the companion metrics visible so binary, ordinal, regression, and multiclass cards are not compared through one flattened score.

Source podium · · 10 families
Gold
PsyProxy
0.307
Silver
TextDescriptives
0.163
Bronze
TAASSC
0.145
Model-family mix
PsyProxy · 4Baseline · 15Lexicon · 2OpenAI / LLM · 3Topic model · 2
SystemFamilyVariantRMSEMAEPrimary scale
psyproxyPsyProxy — Behavioral Sciences Lens v0.5 · 1000d
PsyProxypermissive0.3070.310.31
psyproxyPsyProxy — Social Economics Lens v0.5 · 1000d
PsyProxypermissive0.3010.300.30
psyproxyPsyProxy — Health Lens v0.9 · 1100d
PsyProxypermissive0.2590.260.26
psyproxyPsyProxy — Technology Lens v0.5 · 800d
PsyProxypermissive0.2480.250.25
baselineTextDescriptives
Baselinepermissive0.1630.160.16
baselineTool for the Automatic Analysis of Syntactic Sophistication and Complexity (TAASSC)
Baselinepermissive0.1450.140.14
lexLinguistic Inquiry and Word Count (LIWC)
Lexiconpermissive0.1230.120.12
llmOpenAI Model gpt-5-nano
OpenAI / LLMpermissive0.1010.100.10
baselineTool for the Automatic Analysis of Lexical Sophistication (TAALES)
Baselinepermissive0.0930.090.09
baselineTool for the Automatic Analysis of Cohesion (TAACO)
Baselinepermissive0.0800.080.08
llmOpenAI Model gpt-4o-mini
OpenAI / LLMpermissive0.0800.080.08
llmOpenAI Model gpt-4.1-nano
OpenAI / LLMpermissive0.0650.070.07
topicBERTopic
Topic modelpermissive0.0500.050.05
baselineEmpath
Baselinepermissive0.0110.010.01
topicHierarchical Dirichlet Process (tomotopy HDP)
Topic modelpermissive0.0020.000.00
lexValence Aware Dictionary and sEntiment Reasoner (VADER)
Lexiconpermissive0.0000.000.00
baselineDruglib drug reviews ordinal · via best lens
Baselinepermissive
baselineIMDB movie reviews (ACL) binary · via best lens
Baselinepermissive
baselineDisneyland TripAdvisor reviews continuous · via best lens
Baselinepermissive
baselineAmazon Video-Games reviews ordinal · via best lens
Baselinepermissive
baselineAmazon Video-Games reviews continuous · via best lens
Baselinepermissive
baselineDouban movie reviews (Chinese) ordinal · via best lens
Baselinepermissive
baselineAmazon Video-Games reviews binary · via best lens
Baselinepermissive
baselineDisneyland TripAdvisor reviews binary · via best lens
Baselinepermissive
baselineSentiment140 tweets binary · via best lens
Baselinepermissive
baselineLIAR fact-check statements ordinal · via best lens
Baselinepermissive