label and checkboxes placement changes

.gitignore

@@ -168,4 +168,5 @@ google-cloud-cli-469.0.0-linux-x86_64.tar.gz
 /backend/src/chunks
 /backend/merged_files
 /backend/chunks
-google-cloud-cli-479.0.0-linux-x86_64.tar.gz
+google-cloud-cli-linux-x86_64.tar.gz
+.vennv

backend/score.py

@@ -241,20 +241,20 @@ async def post_processing(uri=Form(), userName=Form(), password=Form(), database
         graph = create_graph_database_connection(uri, userName, password, database)
         tasks = set(map(str.strip, json.loads(tasks)))
 
-        if "update_similarity_graph" in tasks:
+        if "materialize_text_chunk_similarities" in tasks:
             await asyncio.to_thread(update_graph, graph)
-            josn_obj = {'api_name': 'post_processing/update_similarity_graph', 'db_url': uri, 'logging_time': formatted_time(datetime.now(timezone.utc))}
+            josn_obj = {'api_name': 'post_processing/materialize_text_chunk_similarities', 'db_url': uri, 'logging_time': formatted_time(datetime.now(timezone.utc))}
             logger.log_struct(josn_obj)
             logging.info(f'Updated KNN Graph')
-        if "create_fulltext_index" in tasks:
+        if "enable_hybrid_search_and_fulltext_search_in_bloom" in tasks:
             await asyncio.to_thread(create_fulltext, uri=uri, username=userName, password=password, database=database,type="entities")
             await asyncio.to_thread(create_fulltext, uri=uri, username=userName, password=password, database=database,type="keyword")
-            josn_obj = {'api_name': 'post_processing/create_fulltext_index', 'db_url': uri, 'logging_time': formatted_time(datetime.now(timezone.utc))}
+            josn_obj = {'api_name': 'post_processing/enable_hybrid_search_and_fulltext_search_in_bloom', 'db_url': uri, 'logging_time': formatted_time(datetime.now(timezone.utc))}
             logger.log_struct(josn_obj)
             logging.info(f'Full Text index created')
-        if os.environ.get('ENTITY_EMBEDDING','False').upper()=="TRUE" and "create_entity_embedding" in tasks:
+        if os.environ.get('ENTITY_EMBEDDING','False').upper()=="TRUE" and "materialize_entity_similarities" in tasks:
             await asyncio.to_thread(create_entity_embedding, graph)
-            josn_obj = {'api_name': 'post_processing/create_entity_embedding', 'db_url': uri, 'logging_time': formatted_time(datetime.now(timezone.utc))}
+            josn_obj = {'api_name': 'post_processing/materialize_entity_similarities', 'db_url': uri, 'logging_time': formatted_time(datetime.now(timezone.utc))}
             logger.log_struct(josn_obj)
             logging.info(f'Entity Embeddings created')
         return create_api_response('Success', message='All tasks completed successfully')

frontend/src/components/FileTable.tsx

@@ -358,7 +358,9 @@ const FileTable = forwardRef<ChildRef, FileTableProps>((props, ref) => {
               ...Array.from(new Set(filesData.map((f) => f.fileSource))).map((t) => {
                 return {
                   title: (
-                    <span className={`${t === fileSourceFilter ? 'n-bg-palette-primary-bg-selected' : ''} p-2`}>{t}</span>
+                    <span className={`${t === fileSourceFilter ? 'n-bg-palette-primary-bg-selected' : ''} p-2`}>
+                      {t}
+                    </span>
                   ),
                   onClick: () => {
                     setFileSourceFilter(t as string);
@@ -579,14 +581,14 @@ const FileTable = forwardRef<ChildRef, FileTableProps>((props, ref) => {
                     item?.fileSource === 's3 bucket' && localStorage.getItem('accesskey') === item?.awsAccessKeyId
                       ? item?.status
                       : item?.fileSource === 'local file'
-                        ? item?.status
-                        : item?.status === 'Completed' || item.status === 'Failed'
-                          ? item?.status
-                          : item?.fileSource == 'Wikipedia' ||
-                            item?.fileSource == 'youtube' ||
-                            item?.fileSource == 'gcs bucket'
-                            ? item?.status
-                            : 'N/A',
+                      ? item?.status
+                      : item?.status === 'Completed' || item.status === 'Failed'
+                      ? item?.status
+                      : item?.fileSource == 'Wikipedia' ||
+                        item?.fileSource == 'youtube' ||
+                        item?.fileSource == 'gcs bucket'
+                      ? item?.status
+                      : 'N/A',
                   model: item?.model ?? model,
                   id: uuidv4(),
                   source_url: item?.url != 'None' && item?.url != '' ? item.url : '',
@@ -599,8 +601,8 @@ const FileTable = forwardRef<ChildRef, FileTableProps>((props, ref) => {
                   language: item?.language ?? '',
                   processingProgress:
                     item?.processed_chunk != undefined &&
-                      item?.total_chunks != undefined &&
-                      !isNaN(Math.floor((item?.processed_chunk / item?.total_chunks) * 100))
+                    item?.total_chunks != undefined &&
+                    !isNaN(Math.floor((item?.processed_chunk / item?.total_chunks) * 100))
                       ? Math.floor((item?.processed_chunk / item?.total_chunks) * 100)
                       : undefined,
                   // total_pages: item?.total_pages ?? 0,

frontend/src/components/Popups/GraphEnhancementDialog/PostProcessingCheckList/index.tsx

@@ -1,5 +1,5 @@
 import { Box, Checkbox, Flex, Typography, useMediaQuery } from '@neo4j-ndl/react';
-import { taskParam } from '../../../../utils/Constants';
+import { POST_PROCESSING_JOBS } from '../../../../utils/Constants';
 import { capitalize } from '../../../../utils/Utils';
 import { useFileContext } from '../../../../context/UsersFiles';
 import { tokens } from '@neo4j-ndl/base';
@@ -17,67 +17,33 @@ export default function PostProcessingCheckList() {
               These options allow you to fine-tune your knowledge graph for improved performance and deeper analysis
             </Typography>
           </Flex>
-          <Flex justifyContent='space-between' flexDirection='column'>
-            <Flex>
-              <Typography variant={tablet ? 'subheading-small' : 'subheading-medium'}>
-                Update Similarity Graph :
-              </Typography>
-              <Typography variant={tablet ? 'body-small' : 'body-medium'}>
-                This option refines the connections between different pieces of information (chunks) within your
-                knowledge graph. By leveraging a k-nearest neighbor algorithm with a similarity threshold (KNN_MIN_SCORE
-                of 0.8), this process identifies and links chunks with high semantic similarity. This results in a more
-                interconnected and insightful knowledge representation, enabling more accurate and relevant search
-                results.
-              </Typography>
-            </Flex>
-            <Flex>
-              <Typography variant={tablet ? 'subheading-small' : 'subheading-medium'}>
-                Create Fulltext Index :
-              </Typography>
-              <Typography variant={tablet ? 'body-small' : 'body-medium'}>
-                This option optimizes search capabilities within your knowledge graph. It rebuilds the full-text index
-                on database labels, ensuring faster and more efficient retrieval of information. This is particularly
-                beneficial for large knowledge graphs, as it significantly speeds up keyword-based searches and improves
-                overall query performance..
-              </Typography>
-            </Flex>
-            <Flex>
-              <Typography variant={tablet ? 'subheading-small' : 'subheading-medium'}>
-                Create Entity Embeddings :
-              </Typography>
-              <Typography variant={tablet ? 'body-small' : 'body-medium'}>
-                Enhances entity analysis by generating numerical representations (embeddings) that capture their
-                semantic meaning. This facilitates tasks like clustering similar entities, identifying duplicates, and
-                performing similarity-based searches.
-              </Typography>
-            </Flex>
+          <Flex justifyContent='space-between' flexDirection='column' gap='6'>
+            {POST_PROCESSING_JOBS.map((job, idx) => (
+              <Flex key={`${job.title}${idx}`}>
+                <Checkbox
+                  label={
+                    <Typography variant='label'>
+                      {job.title
+                        .split('_')
+                        .map((s) => capitalize(s))
+                        .join(' ')}
+                    </Typography>
+                  }
+                  checked={postProcessingTasks.includes(job.title)}
+                  onChange={(e) => {
+                    if (e.target.checked) {
+                      setPostProcessingTasks((prev) => [...prev, job.title]);
+                    } else {
+                      setPostProcessingTasks((prev) => prev.filter((s) => s !== job.title));
+                    }
+                  }}
+                ></Checkbox>
+                <Typography variant={tablet ? 'body-small' : 'body-medium'}>{job.description}</Typography>
+              </Flex>
+            ))}
           </Flex>
         </Flex>
       </div>
-      <Flex flexDirection={tablet ? 'row' : 'column'}>
-        {taskParam.map((task, index) => (
-          <Box key={index}>
-            <Checkbox
-              label={
-                <Typography variant={tablet ? 'subheading-medium' : 'subheading-large'}>
-                  {task
-                    .split('_')
-                    .map((s) => capitalize(s))
-                    .join(' ')}
-                </Typography>
-              }
-              checked={postProcessingTasks.includes(task)}
-              onChange={(e) => {
-                if (e.target.checked) {
-                  setPostProcessingTasks((prev) => [...prev, task]);
-                } else {
-                  setPostProcessingTasks((prev) => prev.filter((s) => s !== task));
-                }
-              }}
-            ></Checkbox>
-          </Box>
-        ))}
-      </Flex>
     </Flex>
   );
 }

frontend/src/utils/Constants.ts

@@ -59,7 +59,7 @@ export const defaultLLM = llms?.includes('openai-gpt-4o-mini')
 export const chatModes =
   process.env?.CHAT_MODES?.trim() != ''
     ? process.env.CHAT_MODES?.split(',')
-    : ['vector', 'graph', 'graph+vector', 'hybrid','hybrid+graph'];
+    : ['vector', 'graph', 'graph+vector', 'hybrid', 'hybrid+graph'];
 export const chunkSize = process.env.CHUNK_SIZE ? parseInt(process.env.CHUNK_SIZE) : 1 * 1024 * 1024;
 export const timeperpage = process.env.TIME_PER_PAGE ? parseInt(process.env.TIME_PER_PAGE) : 50;
 export const timePerByte = 0.2;
@@ -160,7 +160,34 @@ export const buttonCaptions = {
   ask: 'Ask',
 };
 
-export const taskParam: string[] = ['update_similarity_graph', 'create_fulltext_index', 'create_entity_embedding'];
+export const taskParam: string[] = [
+  'materialize_text_chunk_similarities',
+  'enable_hybrid_search_and_fulltext_search_in_bloom',
+  'materialize_entity_similarities',
+];
+export const POST_PROCESSING_JOBS: { title: string; description: string }[] = [
+  {
+    title: 'materialize_text_chunk_similarities',
+    description: `This option refines the connections between different pieces of information (chunks) within your
+                knowledge graph. By leveraging a k-nearest neighbor algorithm with a similarity threshold (KNN_MIN_SCORE
+                of 0.8), this process identifies and links chunks with high semantic similarity. This results in a more
+                interconnected and insightful knowledge representation, enabling more accurate and relevant search
+                results.`,
+  },
+  {
+    title: 'enable_hybrid_search_and_fulltext_search_in_bloom',
+    description: `This option optimizes search capabilities within your knowledge graph. It rebuilds the full-text index
+                on database labels, ensuring faster and more efficient retrieval of information. This is particularly
+                beneficial for large knowledge graphs, as it significantly speeds up keyword-based searches and improves
+                overall query performance.`,
+  },
+  {
+    title: 'materialize_entity_similarities',
+    description: `Enhances entity analysis by generating numerical representations (embeddings) that capture their
+                semantic meaning. This facilitates tasks like clustering similar entities, identifying duplicates, and
+                performing similarity-based searches.`,
+  },
+];
 
 export const nvlOptions: NvlOptions = {
   allowDynamicMinZoom: true,

frontend/src/utils/Utils.ts

@@ -170,16 +170,14 @@ export const filterData = (
   allNodes: ExtendedNode[],
   allRelationships: Relationship[],
   scheme: Scheme
- ) => {
+) => {
   let filteredNodes: ExtendedNode[] = [];
   let filteredRelations: Relationship[] = [];
   let filteredScheme: Scheme = {};
   const entityTypes = Object.keys(scheme).filter((type) => type !== 'Document' && type !== 'Chunk');
   if (graphType.includes('DocumentChunk') && !graphType.includes('Entities')) {
     // Document + Chunk
-    filteredNodes = allNodes.filter(
-      (node) => node.labels.includes('Document') || node.labels.includes('Chunk')
-    );
+    filteredNodes = allNodes.filter((node) => node.labels.includes('Document') || node.labels.includes('Chunk'));
     const nodeIds = new Set(filteredNodes.map((node) => node.id));
     filteredRelations = allRelationships.filter(
       (rel) =>
@@ -190,9 +188,7 @@ export const filterData = (
     filteredScheme = { Document: scheme.Document, Chunk: scheme.Chunk };
   } else if (graphType.includes('Entities') && !graphType.includes('DocumentChunk')) {
     // Only Entity
-    const entityNodes = allNodes.filter(
-      (node) => !node.labels.includes('Document') && !node.labels.includes('Chunk')
-    );
+    const entityNodes = allNodes.filter((node) => !node.labels.includes('Document') && !node.labels.includes('Chunk'));
     filteredNodes = entityNodes ? entityNodes : [];
     const nodeIds = new Set(filteredNodes.map((node) => node.id));
     filteredRelations = allRelationships.filter(
@@ -209,7 +205,7 @@ export const filterData = (
     filteredScheme = scheme;
   }
   return { filteredNodes, filteredRelations, filteredScheme };
- };
+};
 
 export const getDateTime = () => {
   const date = new Date();