{"id":"W4407061858","doi":"10.1016/j.ecoser.2025.101702","title":"Using georeferenced text from social media to map the cultural ecosystem services of freshwater ecosystems","year":2025,"lang":"en","type":"article","venue":"Ecosystem Services","topic":"Land Use and Ecosystem Services","field":"Environmental Science","cited_by":10,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"H2020 European Research Council; Horizon 2020; Generalitat de Catalunya; Agència Catalana de l'Aigua; Horizon 2020 Framework Programme; Canadian Institute for Advanced Research","keywords":"Georeference; Ecosystem services; Ecosystem; Social media; Geography; Environmental resource management; Ecology; Environmental science; World Wide Web; Computer science; Physical geography; Biology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.0007543144,0.0006043547,0.001003367,0.0001534661,0.0007062576,0.0003663479,0.002139335,0.0003293376,0.002234339],"category_scores_gemma":[0.000005788185,0.0003792543,0.0002661306,0.0007882567,0.00002059003,0.0007589078,0.0009783227,0.0002165326,0.001957767],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003131461,"about_ca_system_score_gemma":0.00003540068,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.02286785,"about_ca_topic_score_gemma":0.6636458,"domain_scores_codex":[0.9956687,0.0004438873,0.001334941,0.0009329107,0.0008115444,0.0008080315],"domain_scores_gemma":[0.9978142,0.0002970262,0.0006642601,0.0009030162,0.0001095918,0.0002118642],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0009240378,0.0007220433,0.7881989,0.02274764,0.002506715,0.00007607653,0.0947904,0.01454519,0.0625862,0.001205129,0.009181074,0.002516512],"study_design_scores_gemma":[0.006207677,0.0002841727,0.1066711,0.01306019,0.001686671,0.00005486974,0.09148857,0.2687872,0.03799711,0.002838819,0.4661565,0.004767158],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9924328,0.0003860114,0.00002164048,0.0005929799,0.002024243,0.0009563897,0.001098388,0.0001500137,0.002337574],"genre_scores_gemma":[0.9983823,0.00001420332,0.0002422332,0.0004060302,0.0005384263,0.0001097708,0.0002007957,0.00004605862,0.00006014618],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6815279,"threshold_uncertainty_score":0.9998659,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01588190640631932,"score_gpt":0.2386024165948012,"score_spread":0.2227205101884819,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}