Ggthjy

Is there some relative to Dutch word "kijken" in German?

Can a dragon be stuck looking like a human?

Can we use the stored gravitational potential energy of a building to produce power?

How to prevent users from executing commands through browser URL

Why does String.replaceAll() work differently in Java 8 from Java 9?

insert EOF statement before the last line of file

Contest math problem about crossing out numbers in the table

What flying insects could re-enter the Earth's atmosphere from space without burning up?

Using only 1s, make 29 with the minimum number of digits

Why zero tolerance on nudity in space?

How to prevent cleaner from hanging my lock screen in Ubuntu 16.04

Compress command output by piping to bzip2

Checking for the existence of multiple directories

What to do when being responsible for data protection in your lab, yet advice is ignored?

How to deal with an incendiary email that was recalled

Why did other German political parties disband so fast when Hitler was appointed chancellor?

Chess tournament winning streaks

What does Cypher mean when he says Neo is "gonna pop"?

Process to change collation on a database

Book where aliens are selecting humans for food consumption

How can animals be objects of ethics without being subjects as well?

Broken patches on a road

What is this metal M-shaped device for?

Cryptic with missing capitals

Anomaly detection in structured textual data

anomaly detection alert systemClustering Crime Data which has {latitute, longitude, crime-type} tuplesAnomaly detection for transaction dataNotion of cluster centers and cluster comparison in Density Based AlgorithmsKDE on TF-IDF - sensitive bandwidthRight Way to Input Text Data in Keras Auto Encoderhow to compare different sets of time series dataIs my data good for (DBSCAN) clustering?Anomaly detection using clustering of highly correlated Categorical datamultivariate clustering, dimensionality reduction and data scalling for regression

0

$begingroup$

Pls refer screenshot for sample data. As can be seen most of the fields in data are textual and highly correlated but each row has unique values and hence won't be right to call it categorical. I tried to break down column 1 in two tokens to ABC and X1 and tried KModes clustering where number of clusters = unique values in column1. However each cluster do not have equal density hence some bad data with high density is classified as normal and good data with low density is marked anomalous.
I want to have unsupervised algo where I can force it to use Key1+Key2 as composite primary key for clustering. Algo should then ensure data follows some pattern in each cluster e.g. (alias1, alias2 moves in unison. Multipliers are in close range.)
Kindly suggest what would be the best anomaly detection algo and how to approach this problem.

clustering text-mining anomaly-detection

share

asked 5 mins ago

viral kapadiaviral kapadia

61

$endgroup$

add a comment | 

0

$begingroup$

Pls refer screenshot for sample data. As can be seen most of the fields in data are textual and highly correlated but each row has unique values and hence won't be right to call it categorical. I tried to break down column 1 in two tokens to ABC and X1 and tried KModes clustering where number of clusters = unique values in column1. However each cluster do not have equal density hence some bad data with high density is classified as normal and good data with low density is marked anomalous.
I want to have unsupervised algo where I can force it to use Key1+Key2 as composite primary key for clustering. Algo should then ensure data follows some pattern in each cluster e.g. (alias1, alias2 moves in unison. Multipliers are in close range.)
Kindly suggest what would be the best anomaly detection algo and how to approach this problem.

clustering text-mining anomaly-detection

share

asked 5 mins ago

viral kapadiaviral kapadia

61

$endgroup$

add a comment | 

$begingroup$

Pls refer screenshot for sample data. As can be seen most of the fields in data are textual and highly correlated but each row has unique values and hence won't be right to call it categorical. I tried to break down column 1 in two tokens to ABC and X1 and tried KModes clustering where number of clusters = unique values in column1. However each cluster do not have equal density hence some bad data with high density is classified as normal and good data with low density is marked anomalous.
I want to have unsupervised algo where I can force it to use Key1+Key2 as composite primary key for clustering. Algo should then ensure data follows some pattern in each cluster e.g. (alias1, alias2 moves in unison. Multipliers are in close range.)
Kindly suggest what would be the best anomaly detection algo and how to approach this problem.

clustering text-mining anomaly-detection

share

asked 5 mins ago

viral kapadiaviral kapadia

61

$endgroup$

share

asked 5 mins ago

viral kapadiaviral kapadia

61

share

asked 5 mins ago

viral kapadiaviral kapadia

61

asked 5 mins ago

viral kapadiaviral kapadia

61

asked 5 mins ago

viral kapadiaviral kapadia

61