e-LICO Recommender Systems2012-01-27T13:10:41+00:00/groups/7872012-05-17T11:22:10+00:00Matej MihelÄić shared Hybrid recommender RapidMiner and extension operators This pack contains workflow and data for hybrid recommender that is created by combining Recommender extension and RapidMiner built in operators. urn:uuid:4f45abbd-2906-4edd-bae3-a9dbd022afc3Matej MihelÄić2012-05-17T11:17:45+00:00Matej MihelÄić shared Hybrid recommendation system This is one hybrid recommendation system combining linear regression recommender, created using RapidMiner core operators, and Recommender extension multiple collaborative filtering and attribute based operators.urn:uuid:c3e5aacb-eb6a-42e9-98aa-9f2eccee59dbMatej MihelÄić2012-05-17T11:06:11+00:00Matej MihelÄić shared MovieLens data for hybrid recommenders This is MovieLens data set containing user,item, ratings and item attributes enabling hybrid recommendation system testing. urn:uuid:70f79c91-f057-4fd1-b2d4-86793720bda0Matej MihelÄić2012-02-10T00:54:06+00:00Matej MihelÄić shared Parameter optimizationThis is a parameter optimization workflow for rating prediction recommendation operators.urn:uuid:458acc65-0634-4c2c-82bc-af9521977213Matej MihelÄić2012-01-31T20:40:15+00:00Lawrynka shared Digital Multimedia Repositories Ontology (DMRO) and KB with RDF version of Videolectures.net dataset For the information on the ontology see : http://www.e-lico.eu/?q=node/288 For the information on the original dataset see: http://www.ecmlpkdd2011.org/challenge.php The ontology and KB files are zipped into one file. urn:uuid:e2cf2cc1-cd89-4a5b-8b5a-4ab20c4b8c0eLawrynka2012-01-31T16:01:22+00:00Matej MihelÄić shared Experimentation through repository accessThis workflow reads train/test dataset from a specified RapidMiner repository and tests selected operator on that datasets. Only datasets specified with a proper regular expression are considered. Train and test data filenames must correspond e.g (train1, test1). Informations about training and testing data, performanse measures of a selected operator are stored as an Excel file. Note: Train/test file names should not be contained in the repository path. E.g training/train is not a god path, while data/train is.urn:uuid:568ab2df-f460-4999-997e-f1d436191c80Matej MihelÄić2012-01-30T15:43:00+00:00tomS shared Transforming user/item description datasets into binomial format (RM recommenders)This workflow provides transformation of an user/item description attribute set, into a format required by attribute based k-NN operators of the Recommender extension. See: http://zel.irb.hr/wiki/lib/exe/fetch.php?media=del:projects:elico:recsys_manual_v1.1.pdf to learn about formats of datasets required by Recommender extension.urn:uuid:3d7d2416-6e3a-4070-88bd-50bb950f794ctomS2012-01-30T13:16:01+00:00Matej MihelÄić shared Recommender workflowThis is a main online update experimentation workflow. It consists of three Execute Process operators. First operator executes model training workflow. Second operator executes online updates workflow for multiple query update sets. The last operator executes performance testing and comparison workflow. Final performance results are saved in an Excel file.urn:uuid:7e738d1c-26df-4c1b-877d-19654ec7567cMatej MihelÄić2012-01-30T13:15:44+00:00Matej MihelÄić shared Model saving workflowThis workflow trains and saves a model for a selected item recommendation operator.urn:uuid:3a31998b-5f34-45ab-b878-f6fc8f75b820Matej MihelÄić2012-01-30T13:06:01+00:00Matej MihelÄić shared Model testing workflowThis workflow measures performance of three models. Model learned on train data and upgraded using online model updates. Model learned on train data + all query update sets. Model learned on train data only.urn:uuid:f8d37cc9-ff68-451c-a299-ef9b1dea0ac2Matej MihelÄić2012-01-30T00:15:42+00:00Lawrynka shared Semantic clustering (with alpha-clustering) of SPARQL query results over RDF version of videolectures.net datasetThe workflow uses RapidMiner extension named RMonto ( http://semantic.cs.put.poznan.pl/RMonto/ ) to perform clustering of SPARQL query results based on chosen semantic similarity measure. The measure used in this particualr workflow is a kernel that exploits membership of clustered individuals to OWL classes from a background ontology ("Epistemic" kernel from [1]). Since the semantics of the backgound ontology is used in this way, we use the name "semantic clustering". This particular kernel is based on the commitee of features (ontology classes) that are specified in the paremeter of "Create TBox features" operator. The SPARQL query is entered in a parameter of "SPARQL selector" operator. The clustering operator (alpha-clustering) allows to specify which of the query variables are to be us …urn:uuid:5e8f8605-1e41-4d72-a320-eed58b15dc6dLawrynka2012-01-29T22:30:37+00:00Lawrynka shared Semantic clustering (with k-medoids) of SPARQL query results over RDF version of videolectures.net datasetThe workflow uses RapidMiner extension named RMonto ( http://semantic.cs.put.poznan.pl/RMonto/ ) to perform clustering of SPARQL query results based on chosen semantic similarity measure. Since the semantics of the backgound ontology is used in this way, we use the name "semantic clustering". The SPARQL query is entered in a parameter of "SPARQL selector" operator. The clustering operator (k-medoids) allows to specify which of the query variables are to be used as clustering criteria. If more than one variable is used than the results are clustered such that kind of multifaceted hierarchy is computed over them, which is dynamically determined. More on such functionality may be found in [1]. The parameters of the operator allow to choose URI attributes (query variables that are bound to OWL …urn:uuid:bcbe7c2f-16cf-466c-8659-3f7e26c925b6Lawrynka