SPAM
- class PAMI.sequentialPatternMining.basic.SPAM.SPAM(iFile, minSup, sep='\t')[source]
Bases:
_sequentialPatterns- Description:
SPAM is one of the fundamental algorithm to discover sequential frequent patterns in a transactional database. This program employs SPAM property (or downward closure property) to reduce the search space effectively. This algorithm employs breadth-first search technique to find the complete set of frequent patterns in a sequential database.
- Reference:
Ayres, J. Gehrke, T.Yiu, and J. Flannick. Sequential Pattern Mining Using Bitmaps. In Proceedings of the Eighth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. Edmonton, Alberta, Canada, July 2002.
- Parameters:
iFile – str : Name of the Input file to mine complete set of Sequential frequent patterns
oFile – str : Name of the output file to store complete set of Sequential frequent patterns
minSup – float or int or str : minSup measure constraints the minimum number of transactions in a database where a pattern must appear Example: minSup=10 will be treated as integer, while minSup=10.0 will be treated as float
sep – str : This variable is used to distinguish items from one another in a transaction. The default seperator is tab space. However, the users can override their default separator.
- Attributes:
- iFilestr
Input file name or path of the input file
- oFilestr
Name of the output file or the path of output file
- minSupfloat or int or str
The user can specify minSup either in count or proportion of database size. If the program detects the data type of minSup is integer, then it treats minSup is expressed in count. Otherwise, it will be treated as float. Example: minSup=10 will be treated as integer, while minSup=10.0 will be treated as float
- sepstr
This variable is used to distinguish items from one another in a transaction. The default seperator is tab space or . However, the users can override their default separator.
- startTimefloat
To record the start time of the mining process
- endTimefloat
To record the completion time of the mining process
- finalPatternsdict
Storing the complete set of patterns in a dictionary variable
- memoryUSSfloat
To store the total amount of USS memory consumed by the program
- memoryRSSfloat
To store the total amount of RSS memory consumed by the program
- Databaselist
To store the sequences of a database in list
- _idDatabasedict
To store the sequences of a database by bit map
- _maxSeqLen:
the maximum length of subsequence in sequence.
- Methods:
- _creatingItemSets():
Storing the complete sequences of the database/input file in a database variable
- _convert(value):
To convert the user specified minSup value
- make2BitDatabase():
To make 1 length frequent patterns by breadth-first search technique and update Database to sequential database
- DfsPruning(items,sStep,iStep):
the main algorithm of spam. This can search sstep and istep items and find next patterns, its sstep, and its istep. And call this function again by using them. Recursion until there are no more items available for exploration.
- Sstep(s):
To convert bit to ssteo bit.The first time you get 1, you set it to 0 and subsequent ones to 1.(like 010101=>001111, 00001001=>00000111)
- mine()
Mining process will start from here
- getPatterns()
Complete set of patterns will be retrieved with this function
- savePatterns(oFile)
Complete set of frequent patterns will be loaded in to a output file
- getPatternsAsDataFrame()
Complete set of frequent patterns will be loaded in to a dataframe
- getMemoryUSS()
Total amount of USS memory consumed by the mining process will be retrieved from this function
- getMemoryRSS()
Total amount of RSS memory consumed by the mining process will be retrieved from this function
- getRuntime()
Total amount of runtime taken by the mining process will be retrieved from this function
- candidateToFrequent(candidateList)
Generates frequent patterns from the candidate patterns
- frequentToCandidate(frequentList, length)
Generates candidate patterns from the frequent patterns
Executing the code on terminal:
Format: (.venv) $ python3 SPAM.py <inputFile> <outputFile> <minSup> (<separator>) Examples usage: (.venv) $ python3 SPAM.py sampleDB.txt patterns.txt 10.0 .. note:: minSup will be considered in times of minSup and count of database transactions
Sample run of the importing code:
import PAMI.sequentialPatternMining.basic.SPAM as alg
obj = alg.SPAM(iFile, minSup)
obj.mine()
sequentialPatternMining = obj.getPatterns()
print(“Total number of Frequent Patterns:”, len(frequentPatterns))
obj.savePatterns(oFile)
Df = obj.getPatternInDataFrame()
memUSS = obj.getMemoryUSS()
print(“Total Memory in USS:”, memUSS)
memRSS = obj.getMemoryRSS()
print(“Total Memory in RSS”, memRSS)
run = obj.getRuntime()
print(“Total ExecutionTime in seconds:”, run)
Credits:
The complete program was written by Shota Suzuki under the supervision of Professor Rage Uday Kiran.
- DfsPruning(items, sStep, iStep)[source]
the main algorithm of spam. This can search sstep and istep items and find next patterns, its sstep, and its istep. And call this function again by using them. Recursion until there are no more items available for exploration.
- Attributes:
- itemsstr
The pattrens I got before
- sSteplist
Items presumed to have “sstep” relationship with “items”.(sstep is What appears later like a-b and a-c)
- iSteplist
Items presumed to have “istep” relationship with “items”(istep is What appears in same time like ab and ac)
- Sstep(s)[source]
To convert bit to Sstep bit.The first time you get 1, you set it to 0 and subsequent ones to 1.(like 010101=>001111, 00001001=>00000111)
- :param s:list
to store each bit sequence
- Returns:
nextS:list to store the bit sequence converted by sstep
- countSup(n)[source]
count support
- :param n:list
to store each bit sequence
- Returns:
count: int support of this list
- getMemoryRSS()[source]
Total amount of RSS memory consumed by the mining process will be retrieved from this function :return: returning RSS memory consumed by the mining process :rtype: float
- getMemoryUSS()[source]
Total amount of USS memory consumed by the mining process will be retrieved from this function :return: returning USS memory consumed by the mining process :rtype: float
- getPatterns()[source]
Function to send the set of frequent patterns after completion of the mining process :return: returning frequent patterns :rtype: dict
- getPatternsAsDataFrame()[source]
Storing final frequent patterns in a dataframe :return: returning frequent patterns in a dataframe :rtype: pd.DataFrame
- getRuntime()[source]
Calculating the total amount of runtime taken by the mining process :return: returning total amount of runtime taken by the mining process :rtype: float
- make2BitDatabase()[source]
To make 1 length frequent patterns by breadth-first search technique and update Database to sequential database