Skip to content

Add complex analyses stats and SPARQL queries (q9-mediators, q9-moderators, q9-mediation-by-category) #10

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
imane19972004 wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Add complex analyses stats and SPARQL queries (q9-mediators, q9-moderators, q9-mediation-by-category) #10

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
b4dfdd9
Add complex analyses stats: types, top mediators, top moderators, VI/…
imane19972004 Jun 5, 2026
f553922
Document complex analyses VI categorization: N.A. bundles are expecte…
imane19972004 Jun 5, 2026
664f358
Add CQs q9-mediators, q9-moderators, q9-mediation-by-category for com…
imane19972004 Jun 5, 2026
faeb29c
Add mediation path stats and mediators by VI category to stats_bdd
imane19972004 Jun 5, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
88 changes: 88 additions & 0 deletions services/SPARQL-Generator/index.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2328,6 +2328,92 @@ WHERE {
BIND(REPLACE(STR(?sportURI), '.*/([^/]+)$', '$1') AS ?sportCategory)
}
ORDER BY ?vi ?vd
`;
break;

case 'q9-mediators':
console.log(" CASE Q9-MEDIATORS DÉTECTÉ: Analyses complexes — tous les médiateurs");
selectedCase = 'q9-mediators - Médiateurs dans les analyses complexes';
expectedResults = '100-150 analyses avec médiateur renseigné';

query = `${prefixes}

SELECT DISTINCT ?vi ?vd ?mediator ?mediatorMeasure ?typeAnalysis ?direction ?analysis
WHERE {
?analysis a iadas:Analysis ;
iadas:complexityOfAnalysis "Complex analyses" ;
iadas:hasMediator ?mediator ;
iadas:mediatorMeasure ?mediatorMeasure ;
iadas:typeOfAnalysis ?typeAnalysis ;
iadas:hasRelation ?relation .
FILTER(?mediator != "N.A.")

?relation iadas:hasIndependentVariable ?variableVI ;
iadas:hasDependentVariable ?variableVD .
?variableVI iadas:variableName ?vi .
?variableVD iadas:variableName ?vd .

OPTIONAL { ?analysis iadas:relationDirection ?direction }
}
ORDER BY ?mediator ?vi
`;
break;

case 'q9-moderators':
console.log(" CASE Q9-MODERATORS DÉTECTÉ: Analyses complexes — tous les modérateurs");
selectedCase = 'q9-moderators - Modérateurs dans les analyses complexes';
expectedResults = '25-40 analyses avec modérateur renseigné';

query = `${prefixes}

SELECT DISTINCT ?vi ?vd ?moderator ?moderatorMeasure ?typeAnalysis ?direction ?analysis
WHERE {
?analysis a iadas:Analysis ;
iadas:complexityOfAnalysis "Complex analyses" ;
iadas:hasModerator ?moderator ;
iadas:moderatorMeasure ?moderatorMeasure ;
iadas:typeOfAnalysis ?typeAnalysis ;
iadas:hasRelation ?relation .
FILTER(?moderator != "N.A.")

?relation iadas:hasIndependentVariable ?variableVI ;
iadas:hasDependentVariable ?variableVD .
?variableVI iadas:variableName ?vi .
?variableVD iadas:variableName ?vd .

OPTIONAL { ?analysis iadas:relationDirection ?direction }
}
ORDER BY ?moderator ?vi
`;
break;

case 'q9-mediation-by-category':
console.log(" CASE Q9-MEDIATION-BY-CATEGORY DÉTECTÉ: Médiateurs par catégorie SKOS de VI");
selectedCase = 'q9-mediation-by-category - Chemins de médiation par catégorie';
expectedResults = '60-80 chemins de médiation avec catégorie VI identifiée';

query = `${prefixes}

SELECT DISTINCT ?categoryVI ?vi ?mediator ?vd ?typeAnalysis ?analysis
WHERE {
?analysis a iadas:Analysis ;
iadas:complexityOfAnalysis "Complex analyses" ;
iadas:hasMediator ?mediator ;
iadas:typeOfAnalysis ?typeAnalysis ;
iadas:hasRelation ?relation .
FILTER(?mediator != "N.A.")

?relation iadas:hasIndependentVariable ?variableVI ;
iadas:hasDependentVariable ?variableVD .
?variableVI iadas:variableName ?vi ;
iadas:refersToVariable ?viConcept .
?viConcept skos:broader+ ?top .
?top skos:prefLabel ?categoryVI .
FILTER NOT EXISTS { ?top skos:broader ?x . FILTER(CONTAINS(STR(?x), 'ACAD-vocab')) }

?variableVD iadas:variableName ?vd .
}
ORDER BY ?categoryVI ?mediator
`;
break;

Expand All @@ -2341,6 +2427,8 @@ ORDER BY ?vi ?vd
console.error(" - q6-female, q6-male, q6-mixed");
console.error(" - q7-volleyball-men-21");
console.error(" - q8-individual, q8-team, q8-mixed, q8-aesthetic");
console.error(" [Analyses complexes]");
console.error(" - q9-mediators, q9-moderators, q9-mediation-by-category");
console.log("🔧 Utilisation d'une requête par défaut...");

selectedCase = 'DEFAULT - Requête générale de secours';
Expand Down
143 changes: 142 additions & 1 deletion stats_bdd.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,7 +31,8 @@ def count(sparql):
PREFIX dct: <http://purl.org/dc/terms/>
"""

FILTER_SIMPLES = '?c = "Simple analyses" || ?c = "simple analyses"'
FILTER_SIMPLES = '?c = "Simple analyses" || ?c = "simple analyses"'
FILTER_COMPLEXES = '?c = "Complex analyses"'

print("=" * 55)
print(f" STATS BDD IADAS — {datetime.now().strftime('%d/%m/%Y %H:%M')}")
Expand Down Expand Up @@ -273,4 +274,144 @@ def count(sparql):
print(f" {'Groupes article+VI+VD (sous-groupes OK)':<35} {nb_groupes_rep:>6}")
print(f" {'Vrais doublons (+ direction+valeur+n=)':<35} {nb_vrais_doublons:>6} [OK]")

# --- Analyses complexes : vue d'ensemble ---
print("\n ANALYSES COMPLEXES — VUE D'ENSEMBLE")
print(" " + "-" * 40)
cx_total = count(f"{PREFIX} SELECT (COUNT(DISTINCT ?a) AS ?n) WHERE {{ ?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c . FILTER({FILTER_COMPLEXES}) }}")
cx_med = count(f"{PREFIX} SELECT (COUNT(DISTINCT ?a) AS ?n) WHERE {{ ?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ; iadas:hasMediator ?m . FILTER({FILTER_COMPLEXES}) FILTER(?m != 'N.A.') }}")
cx_mod = count(f"{PREFIX} SELECT (COUNT(DISTINCT ?a) AS ?n) WHERE {{ ?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ; iadas:hasModerator ?m . FILTER({FILTER_COMPLEXES}) FILTER(?m != 'N.A.') }}")
cx_medmod = count(f"{PREFIX} SELECT (COUNT(DISTINCT ?a) AS ?n) WHERE {{ ?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ; iadas:hasMediator ?med ; iadas:hasModerator ?mod . FILTER({FILTER_COMPLEXES}) FILTER(?med != 'N.A.') FILTER(?mod != 'N.A.') }}")
print(f" {'Total analyses complexes':<35} {cx_total:>6}")
print(f" {'Avec mediateur renseigne':<35} {cx_med:>6}")
print(f" {'Avec moderateur renseigne':<35} {cx_mod:>6}")
print(f" {'Avec mediateur ET moderateur':<35} {cx_medmod:>6}")

# --- Repartition par type d'analyse ---
print("\n TYPES D'ANALYSES COMPLEXES (top 10)")
print(" " + "-" * 40)
d_types = query(f"""
{PREFIX}
SELECT ?type (COUNT(*) AS ?n)
WHERE {{
?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ; iadas:typeOfAnalysis ?type .
FILTER({FILTER_COMPLEXES})
}}
GROUP BY ?type ORDER BY DESC(?n)
""")
for b in d_types["results"]["bindings"][:10]:
print(f" {b['type']['value'][:35]:<35} {b['n']['value']:>6}")

# --- Top mediateurs ---
print("\n TOP MEDIATEURS (analyses complexes)")
print(" " + "-" * 40)
d_meds = query(f"""
{PREFIX}
SELECT ?mediateur (COUNT(DISTINCT ?a) AS ?n)
WHERE {{
?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ; iadas:hasMediator ?mediateur .
FILTER({FILTER_COMPLEXES})
FILTER(?mediateur != "N.A.")
}}
GROUP BY ?mediateur ORDER BY DESC(?n)
""")
for b in d_meds["results"]["bindings"][:10]:
label = b["mediateur"]["value"][:35]
print(f" {label:<35} {b['n']['value']:>6}")

# --- Top moderateurs ---
print("\n TOP MODERATEURS (analyses complexes)")
print(" " + "-" * 40)
d_mods = query(f"""
{PREFIX}
SELECT ?moderateur (COUNT(DISTINCT ?a) AS ?n)
WHERE {{
?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ; iadas:hasModerator ?moderateur .
FILTER({FILTER_COMPLEXES})
FILTER(?moderateur != "N.A.")
}}
GROUP BY ?moderateur ORDER BY DESC(?n)
""")
for b in d_mods["results"]["bindings"][:10]:
label = b["moderateur"]["value"][:35]
print(f" {label:<35} {b['n']['value']:>6}")

# --- Categorisation VI/VD dans les complexes ---
# Note : dans les analyses complexes, une VI peut contenir plusieurs variables
# bundlees (ex: "BMI\nAge\nGender") → refersToVariable = N.A. (normal et intentionnel).
# Seules les analyses a VI unique et bien definie obtiennent un mapping ACAD.
print("\n CATEGORISATION VI/VD (analyses complexes)")
print(" " + "-" * 40)
cx_vi_total = count(f"{PREFIX} SELECT (COUNT(DISTINCT ?v) AS ?n) WHERE {{ ?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ; iadas:hasRelation ?rel . FILTER({FILTER_COMPLEXES}) ?rel iadas:hasIndependentVariable ?v }}")
cx_vi_na = count(f"{PREFIX} SELECT (COUNT(DISTINCT ?v) AS ?n) WHERE {{ ?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ; iadas:hasRelation ?rel . FILTER({FILTER_COMPLEXES}) ?rel iadas:hasIndependentVariable ?v . ?v iadas:refersToVariable ?concept . FILTER(CONTAINS(STR(?concept),'N.A')) }}")
cx_vi_cat = count(f"{PREFIX} SELECT (COUNT(DISTINCT ?v) AS ?n) WHERE {{ ?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ; iadas:hasRelation ?rel . FILTER({FILTER_COMPLEXES}) ?rel iadas:hasIndependentVariable ?v . ?v iadas:refersToVariable ?concept . ?concept skos:broader ?b }}")
cx_vd_total = count(f"{PREFIX} SELECT (COUNT(DISTINCT ?v) AS ?n) WHERE {{ ?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ; iadas:hasRelation ?rel . FILTER({FILTER_COMPLEXES}) ?rel iadas:hasDependentVariable ?v }}")
cx_vd_cat = count(f"{PREFIX} SELECT (COUNT(DISTINCT ?v) AS ?n) WHERE {{ ?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ; iadas:hasRelation ?rel . FILTER({FILTER_COMPLEXES}) ?rel iadas:hasDependentVariable ?v . ?v iadas:refersToVariable ?concept . ?concept skos:broader ?b }}")
# Analyses avec au moins une VI catégorisée (métrique pertinente pour les complexes)
cx_an_with_vi_cat = count(f"{PREFIX} SELECT (COUNT(DISTINCT ?a) AS ?n) WHERE {{ ?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ; iadas:hasRelation ?rel . FILTER({FILTER_COMPLEXES}) ?rel iadas:hasIndependentVariable ?v . ?v iadas:refersToVariable ?concept . ?concept skos:broader ?b }}")
pct_cx_vi = round(cx_vi_cat / cx_vi_total * 100) if cx_vi_total > 0 else 0
pct_cx_vd = round(cx_vd_cat / cx_vd_total * 100) if cx_vd_total > 0 else 0
pct_cx_an = round(cx_an_with_vi_cat / cx_total * 100) if cx_total > 0 else 0
print(f" {'VI total (complexes)':<35} {cx_vi_total:>6}")
print(f" {'VI bundles (N.A. — normal)':<35} {cx_vi_na:>6} (multi-predicteurs)")
print(f" {'VI avec categorie SKOS':<35} {cx_vi_cat:>6} ({pct_cx_vi}%)")
print(f" {'Analyses avec >=1 VI categ.':<35} {cx_an_with_vi_cat:>6} ({pct_cx_an}% des complexes)")
print(f" {'VD total (complexes)':<35} {cx_vd_total:>6}")
print(f" {'VD avec categorie SKOS':<35} {cx_vd_cat:>6} ({pct_cx_vd}%)")

# --- Top chemins de médiation VI → Médiateur → VD ---
print("\n TOP CHEMINS DE MEDIATION (VI --> Mediateur --> VD)")
print(" " + "-" * 40)
d_chemins = query(f"""
{PREFIX}
SELECT ?viNom ?mediateur ?vdNom (COUNT(DISTINCT ?a) AS ?n)
WHERE {{
?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ;
iadas:hasMediator ?mediateur ; iadas:hasRelation ?rel .
FILTER({FILTER_COMPLEXES})
FILTER(?mediateur != "N.A.")
?rel iadas:hasIndependentVariable ?vi ; iadas:hasDependentVariable ?vd .
?vi iadas:variableName ?viNom . ?vd iadas:variableName ?vdNom .
}}
GROUP BY ?viNom ?mediateur ?vdNom
ORDER BY DESC(?n)
""")
for b in d_chemins["results"]["bindings"][:12]:
vi = b["viNom"]["value"].split("\n")[0].strip()[:22]
med = b["mediateur"]["value"][:22]
vd = b["vdNom"]["value"].split("\n")[0].strip()[:18]
n = b["n"]["value"]
print(f" [{vi}] --[{med}]--> [{vd}] (n={n})")

# --- Médiateurs par catégorie de VI (SKOS) ---
print("\n MEDIATEURS PAR CATEGORIE DE VI (SKOS)")
print(" " + "-" * 40)
d_med_cat = query(f"""
{PREFIX}
SELECT ?catVI ?mediateur (COUNT(DISTINCT ?a) AS ?n)
WHERE {{
?a a iadas:Analysis ; iadas:complexityOfAnalysis ?c ;
iadas:hasMediator ?mediateur ; iadas:hasRelation ?rel .
FILTER({FILTER_COMPLEXES})
FILTER(?mediateur != "N.A.")
?rel iadas:hasIndependentVariable ?vi .
?vi iadas:refersToVariable ?concept .
?concept skos:broader+ ?top .
?top skos:prefLabel ?catVI .
FILTER NOT EXISTS {{ ?top skos:broader ?x . FILTER(CONTAINS(STR(?x), 'ACAD-vocab')) }}
}}
GROUP BY ?catVI ?mediateur
ORDER BY ?catVI DESC(?n)
""")
from collections import defaultdict
med_by_cat = defaultdict(list)
for b in d_med_cat["results"]["bindings"]:
cat = b["catVI"]["value"]
med = b["mediateur"]["value"].split("\n")[0].strip()[:35]
n = int(b["n"]["value"])
med_by_cat[cat].append((med, n))
for cat, meds in sorted(med_by_cat.items()):
print(f" {cat[:40]}")
for med, n in meds[:4]:
print(f" --> {med:<35} (n={n})")

print("\n" + "=" * 55)