@inproceedings{Vaarandi2003,
   abstract    = {Clustering; Frequent Patterns; Log Analysis},
   notes       = { Domain of Application Any ascii based log-file Data Model
                  Any type of log data Key Ideas: 1. Mining infrequent patterns
                  is as  as frequent patterns. 2. Log files donot have
                  stucture and thus difficult to apply association rule
                  algorithms for detecting temporal associations. 3. Log file
                  lines can be viewed as points from a categorical data set,
                  since each line can be divided into words, with the n-th word
                  serving as a value for the n-th attribute. 4. One approach to
                  Measuring distances between categorical data : Jaccard
                  Coefficient 5. It is meaningless to discover clusters in high
                  dimensional data space and thus measuring distances. 6.
                  Algorithms for high dimensional data clustering:MAFIA CACTUS
                  PROCLUS Apriori 7. Instead, their approach is density based,
                  where a clustering algorithm tries to identify dense regions
                  in the data space, and forms clusters from those regions. 8.
},
   author      = {Vaarandi, Risto},
   booktitle   = {Proceedings of the 2003 IEEE Workshop on IP Operations and
                  Management IPOM},
   file        = {:media/extstor2/knobase/papers/Vaarandi/Proceedings of the
                  2003 IEEE Workshop on IP Operations and Management
                  IPOM/Vaarandi - A data clustering algorithm for mining
                  patterns from event logs - 2003.pdf:pdf},
   keywords    = {Log Analysis,data clustering,data mining,system monitoring},
   keywords= {Log Analysis},
   pages       = {119--126},
   publisher   = {Citeseer},
   title       = {{A data clustering algorithm for mining patterns from event
                  logs}},
   url         = {http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.111.820\&rep=rep1\&type=pdf},
   year        = {2003}
}

{"_id":{"_str":"5418dbd3ffe14fcc4f000ff1"},"__v":0,"authorIDs":[],"author_short":["Vaarandi, R."],"bibbaseid":"vaarandi-adataclusteringalgorithmforminingpatternsfromeventlogs-2003","bibdata":{"bibtype":"inproceedings","type":"inproceedings","abstract":"Clustering; Frequent Patterns; Log Analysis","notes":"Domain of Application Any ascii based log-file Data Model Any type of log data Key Ideas: 1. Mining infrequent patterns is as as frequent patterns. 2. Log files donot have stucture and thus difficult to apply association rule algorithms for detecting temporal associations. 3. Log file lines can be viewed as points from a categorical data set, since each line can be divided into words, with the n-th word serving as a value for the n-th attribute. 4. One approach to Measuring distances between categorical data : Jaccard Coefficient 5. It is meaningless to discover clusters in high dimensional data space and thus measuring distances. 6. Algorithms for high dimensional data clustering:MAFIA CACTUS PROCLUS Apriori 7. Instead, their approach is density based, where a clustering algorithm tries to identify dense regions in the data space, and forms clusters from those regions. 8. ","author":[{"propositions":[],"lastnames":["Vaarandi"],"firstnames":["Risto"],"suffixes":[]}],"booktitle":"Proceedings of the 2003 IEEE Workshop on IP Operations and Management IPOM","file":":media/extstor2/knobase/papers/Vaarandi/Proceedings of the 2003 IEEE Workshop on IP Operations and Management IPOM/Vaarandi - A data clustering algorithm for mining patterns from event logs - 2003.pdf:pdf","keywords":"Log Analysis","pages":"119--126","publisher":"Citeseer","title":"A data clustering algorithm for mining patterns from event logs","url":"http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.111.820&rep=rep1&type=pdf","year":"2003","bibtex":"@inproceedings{Vaarandi2003,\n abstract = {Clustering; Frequent Patterns; Log Analysis},\n notes = { Domain of Application Any ascii based log-file Data Model\n Any type of log data Key Ideas: 1. Mining infrequent patterns\n is as as frequent patterns. 2. Log files donot have\n stucture and thus difficult to apply association rule\n algorithms for detecting temporal associations. 3. Log file\n lines can be viewed as points from a categorical data set,\n since each line can be divided into words, with the n-th word\n serving as a value for the n-th attribute. 4. One approach to\n Measuring distances between categorical data : Jaccard\n Coefficient 5. It is meaningless to discover clusters in high\n dimensional data space and thus measuring distances. 6.\n Algorithms for high dimensional data clustering:MAFIA CACTUS\n PROCLUS Apriori 7. Instead, their approach is density based,\n where a clustering algorithm tries to identify dense regions\n in the data space, and forms clusters from those regions. 8.\n},\n author = {Vaarandi, Risto},\n booktitle = {Proceedings of the 2003 IEEE Workshop on IP Operations and\n Management IPOM},\n file = {:media/extstor2/knobase/papers/Vaarandi/Proceedings of the\n 2003 IEEE Workshop on IP Operations and Management\n IPOM/Vaarandi - A data clustering algorithm for mining\n patterns from event logs - 2003.pdf:pdf},\n keywords = {Log Analysis,data clustering,data mining,system monitoring},\n keywords= {Log Analysis},\n pages = {119--126},\n publisher = {Citeseer},\n title = {{A data clustering algorithm for mining patterns from event\n logs}},\n url = {http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.111.820\\&rep=rep1\\&type=pdf},\n year = {2003}\n}\n\n\n","author_short":["Vaarandi, R."],"key":"Vaarandi2003","id":"Vaarandi2003","bibbaseid":"vaarandi-adataclusteringalgorithmforminingpatternsfromeventlogs-2003","role":"author","urls":{"Paper":"http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.111.820&rep=rep1&type=pdf"},"keyword":["Log Analysis"],"downloads":0},"bibtype":"inproceedings","biburl":"https://dl.dropboxusercontent.com/u/14215034/bibs/bibs/thesis-bb.bib","creationDate":"2014-09-17T00:54:43.560Z","downloads":0,"keywords":["log analysis"],"search_terms":["data","clustering","algorithm","mining","patterns","event","logs","vaarandi"],"title":"A data clustering algorithm for mining patterns from event logs","year":2003,"dataSources":["bWAYKFgHdvrrBcfMA"]}