Skip to content
No description, website, or topics provided.
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
ANNModel
BioPortalClassLM
BioPortalProperty
BioPortalPropertyOntologyRestricted
FilesUsedToBuildtraining
FrequencyDistributions
MatchClassification
Mesh+Sio Matches
RNNModels
RNNfiles
columnMappingUniqueIndex
propertysearchLOV
training
.gitignore
APIselectionsubject.py
BioPortalClass.py
BioPortalFrequency.py
BioPortalProperty.py
CleanHGNC.py
CleaningCTD.py
DataPreprocessforNNArtificial.py
LICENSE
LoVProperty.py
ModelExploration.py
ModelTraining.py
ModelTrainingRedo.py
NNEmbData.py
NNEmbDataName.py
NNEmbDataSymbol.py
Name Analyses.py
NewData.py
README.md
RdfDatatypeDetection.py
ResultsReportClasses.xlsx
ResultsReportProperties.xlsx
buildfilecolumns.py
categories.txt
col_file.txt
col_fileorg.txt
createfilewithcolumns.py
negative examples.py
redofile.py
testn.txt
testp.txt
your_file.txt
your_file2.txt

README.md

semantic-enhancement

->In order to test column similarity on an already trained model, load the preffered model in the file ModelExplore.py and ask for a specfic column

-> In order to test the classification of the concept recgnition model, load the preffered model in the file ModdelTrainingRedo and follow the instructions inside

-> To run the ANN with an already trained model and test it, load the model in the file DataPreprocessforNNArtificial.py

----------------------------------------------------------Folder structure:------------------------------------------------------------

ANNModel - contains variations of the trained Artificial Neural Network for binary classification of the Gene concept using column names

#############################################################################################

BioPortalClassLM- contains csv files for each dataset with results from using The BioPortal API for class search with parameter require_exact_match={true}

#############################################################################################

BioPortalProperty -contains csv files for each dataset with results from using The BioPortal API for property search with no parameter restrictions

#############################################################################################

BioPortalPropertyOntologyRestricted- contains 3 types of csv files (for each dataset):(i)results from using The BioPortal API for property search with restriction of parameter ontology_list={ontology1,ontology2,..,}(results_dataset_restricted.csv ),(ii)results from using The BioPortal API for property search with restriction of parameter ontology_list={ontology1,ontology2,..,} and require_exact_match={true}(results_dataset_restricted_long.csv), (iii) files with categorized matches for the second option(datasetmatched.xslx)

#############################################################################################

FilesUsedToBuildtraining-contains files that were created in the process of writing complete training files for the Neural Network Embedding Model

#############################################################################################

FrequencyDistributions-contains graphs for complete and restricted ontology frequecy distributions for each dataset(the used threshold for each dataset is in the file name)

#############################################################################################

MatchClassification-files containg: (i)results of the proposed framework(.csv) and (ii) the categorizations of the matches(.xslx)

#############################################################################################

RNNModels-trained models of the Recurrent Neural Network

#############################################################################################

RNNfiles- files with ouput of the RNN

#############################################################################################

training- Training files (negative and positive) used for the Neural Network Embedding Model comming from the following approaches: 1)restart indexing for the same category in another dataset, 2)continue indexing for the same category in another dataset

You can’t perform that action at this time.