fix(clustering)

science-engineering-art · Dec 21, 2022 · 6bb5838 · 6bb5838
1 parent 22fee63
commit 6bb5838
Show file tree

Hide file tree

Showing 4 changed files with 45 additions and 8 deletions.
diff --git a/src/client/src/App.tsx b/src/client/src/App.tsx
@@ -53,12 +53,20 @@ function App() {
           count={100} 
           page={page} 
           onChange={async (_, value) => {
-              setPage(value);
-              await axios.get('http://localhost:8000/search?' +
+              var url = 'http://localhost:8000/search?' +
                 'model='+encodeURIComponent(model)+
                 '&dataset=' + encodeURIComponent(dataset) +
                 '&query=' + encodeURIComponent(sessionStorage['query']) + 
-                '&pag=' + encodeURIComponent(value))
+                '&pag=' + encodeURIComponent(value);
+              console.log(sessionStorage['model'])
+              if (sessionStorage['model'] === 'clustering')
+                url = 'http://localhost:8000/clustering?' +
+                'dataset=' + encodeURIComponent(dataset) +
+                '&query=' + encodeURIComponent(sessionStorage['query']) + 
+                '&cluster=' + encodeURIComponent(value);
+
+                setPage(value);
+              await axios.get(url)
                 .then((resp) => {
                   setDocumentDtos(resp.data.results);
                   setShowResults(true);

diff --git a/src/client/src/components/QueryBar.tsx b/src/client/src/components/QueryBar.tsx
@@ -32,10 +32,19 @@ function QueryBar(props: QueryBarProps) {
             className="searchButton"
             onClick={async () => {
               sessionStorage[ 'query'] = query
-              await axios.get('http://localhost:8000/search?' +
+
+              var url = 'http://localhost:8000/search?' +
                 'model='+encodeURIComponent(sessionStorage['model'])+
                 '&dataset=' + encodeURIComponent(sessionStorage['dataset']) +
-                '&query=' + encodeURIComponent(query))
+                '&query=' + encodeURIComponent(sessionStorage['query']) + 
+                '&pag=' + encodeURIComponent(sessionStorage['pag']);
+              if (sessionStorage['model'] === 'clustering')
+                 url = 'http://localhost:8000/clustering?' +
+                  'dataset=' + encodeURIComponent(sessionStorage['dataset']) +
+                  '&query=' + encodeURIComponent(sessionStorage['query']) + 
+                  '&cluster=' + encodeURIComponent(sessionStorage['pag']);
+
+              await axios.get(url)
                 .then((resp) => {
                   props.setDocumentDtos(resp.data.results);
                   props.setShowResults(true);})

diff --git a/src/main.py b/src/main.py
@@ -8,7 +8,7 @@
 from models.corpus import Corpus
 from models.fuzzy_model import FuzzyModel
 from models.vector_model import VectorModel
-# from models.kmeans_based_model import VectorModelKMEANS
+from models.kmeans_based_model import VectorModelKMEANS
 from models.relevance_feedback import RelevanceFeedback
 import dictdatabase as ddb
 
@@ -99,4 +99,22 @@ async def feedbackController(
             title=doc['title'], author=doc['author'], 
             text=doc['text'], score=tuple[0]))
 
-    return { "results": result }
+    return { "results": result }
+
+@app.get('/clustering')
+async def clusteringController(dataset: str, query: str, cluster: int):
+
+    if type(models['clustering'][dataset]) == str:
+        models['clustering'][dataset] = eval(models['clustering'][dataset])
+    ranking: VectorModelKMEANS = models['clustering'][dataset]
+
+    result = []
+
+    for x, score, doc_id in ranking.searchSplitedByClusters(query)[cluster-1]:
+        print(x, score, doc_id)
+        doc = corpus[dataset].get_doc(doc_id)
+        result.append(DocumentDto(doc_id=doc['doc_id'], 
+            title=doc['title'], author=doc['author'], 
+            text=doc['text'], score=score))
+
+    return {"results": result}
diff --git a/src/models/kmeans_based_model.py b/src/models/kmeans_based_model.py
@@ -82,7 +82,9 @@ def searchSplitedByClusters(self, query : str):
         results_by_cluster = [[] for _ in range(self.noClusters)]
         for score, doc_id in results:
             results_by_cluster[self.kmeans.labels_[self.doc_postion[doc_id]]].append((self.kmeans.labels_[self.doc_postion[doc_id]], score, doc_id))
-
+
+        results_by_cluster = sorted(results_by_cluster, key=lambda x: x[0][1], reverse=True)
+
         return results_by_cluster
 
     def GetQueryVector(idfs, terms, query):