PPT-Clustering CSC 575 Intelligent Information Retrieval

2 Clustering Agenda Clustering Problem and Clustering Applications Clustering Methodologies and Techniques Graphbased clustering methods KMeans and allocationbased methods Hierarchical Agglomerative Clustering. yahoocomqsNTTOdt brPage 6br 40 CGI CSC309 11 40 CGI CSC309 12 AUTHTYPE CONTENTLENGTH CONTENTTYPE GATEWAYINTERFACE PATHINFO PATHTRANSLATED QUERYSTRING REMOTEADDR REMOTEHOST REMOTEIDENT REMOTEUSER REQUESTMETHOD SCRIPTNAME SERVERNAME SERVERPORT SERVER CSC . 575. Intelligent Information Retrieval. 2. Source: . Intel. How much information?. Google: . ~100 . PB a . day; 3+ million servers (15 . Exabytes. stored). Wayback Machine has . ~9 . PB + . 100 . CSC 575. Intelligent Information Retrieval. Intelligent Information Retrieval. 2. Clustering Techniques and IR. Today. Clustering Problem and Applications. Clustering Methodologies and Techniques. Applications of Clustering in IR. CSC 575. Intelligent Information Retrieval. Intelligent Information Retrieval. 2. Indexing. Indexing is the process of transforming items (documents) into a searchable data structure. creation of document surrogates to represent each document. Nouf Aljaffan (C) 2012 - CSC 1201 Course at KSU. Warning. Lab and . tutorial. work requires your attendance. It is . not a group work. therefore you must study before the class time to be able to finish the required assignments or tutorial. . CSC . Bazaar. CSCs can sell products through this online e-commerce portal enabled through . Infibeam. .. Further CSCs can also open its own online store for free and sell local products through. it. The product can be delivered to the customer either at the CSC Location where the order. PAN . Card Service. The UTI ITSL portal for Pan Card Service and the CSC India Online portal have been. integrated to offer seamless delivery of PAN Card services to citizens through CSCs.. Presently only new applications of PAN requests are taken. A new Pan card to rural citizens. Unsupervised . learning. Seeks to organize data . into . “reasonable” . groups. Often based . on some similarity (or distance) measure defined over data . elements. Quantitative characterization may include. Lecture outline. Distance/Similarity between data objects. Data objects as geometric data points. Clustering problems and algorithms . K-means. K-median. K-center. What is clustering?. A . grouping. of data objects such that the objects . Dr. Mary-Angela Papalaskari. Department of Computing Sciences. Villanova University. Course website:. www.csc.villanova.edu/~map/1051/. Some slides in this presentation are adapted from the slides accompanying Java Software Solutions by Lewis & Loftus. Produces a set of . nested clusters . organized as a hierarchical tree. Can be visualized as a . dendrogram. A . tree-like . diagram that records the sequences of merges or splits. Strengths of Hierarchical Clustering. Log. 2. transformation. Row centering and normalization. Filtering. Log. 2. Transformation. Log. 2. -transformation makes sure that the noise is independent of the mean and similar differences have the same meaning along the dynamic range of the values.. Randomization tests. Cluster Validity . All clustering algorithms provided with a set of points output a clustering. How . to evaluate the “goodness” of the resulting clusters?. Tricky because . What is clustering?. Grouping set of documents into subsets or clusters.. The Goal of clustering algorithm is:. To create clusters that are coherent internally, but clearly different from each other.