arulesSequences包做序列模式的关联分析

library(arulesSequences)

tmp_data<-data.frame(item=factor(c("A","B","B","A","B","A","C","A","B","C","B","A","B","A","A","B","A","B")))#必须是factor
tmp_data.tran<-as(tmp_data,"transactions")
transactionInfo(tmp_data.tran)$sequenceID<-c(1,1,1,1,1,2,2,2,2,2,2,3,3,3,4,4,4,4)
transactionInfo(tmp_data.tran)$eventID<-c(10,10,20,30,30,20,20,30,30,30,50,10,30,40,30,30,40,50)
transactionInfo(tmp_data.tran)

#lab1：基本约束support（默认为0.1）
s_result<-cspade(tmp_data.tran,parameter = list(support = 0.75),control = list(verbose = TRUE))
inspect(s_result)
#   items      support
# 1 <{item=A}>    1.00
# 2 <{item=B}>    1.00
# 3 <{item=A},
#    {item=B}>    1.00
# 4 <{item=A,
#     item=B}>    0.75
# 5 <{item=B},
#    {item=B}>    0.75
# 6 <{item=A,
#     item=B},
#    {item=B}>    0.75
# 7 <{item=A},
#    {item=A}>    1.00
# 8 <{item=B},
#    {item=A}>    0.75

#lab2：maxlen约束：每一个序列的event最多只能为n（每一个序列中的“{}”号最多只能为n个）
#英文解释 maxlen: an integer value specifying the maximum number of elements of a sequence (default 10 range > 0).
#element就是指event。所以：sequence包含event（element），event包含item
s_result<-cspade(tmp_data.tran,parameter = list(support = 0.75,maxlen=1),control = list(verbose = TRUE))
inspect(s_result)
#   items      support
# 1 <{item=A}>    1.00
# 2 <{item=B}>    1.00
# 3 <{item=A,
#     item=B}>    0.75

#lab3：maxsize约束：每一个序列的每个event内部最多只能有n个item（每一个序列中的每个{}内部最多有n个items）
#英文解释maxsize: an integer value specifying the maximum number of items of an element of a sequence(default 10, range > 0).
#element就是指event。所以：sequence包含event（element），event包含item
s_result<-cspade(tmp_data.tran,parameter = list(support = 0.75,maxsize=1),control = list(verbose = TRUE))
inspect(s_result)
#   items      support
# 1 <{item=A}>    1.00
# 2 <{item=B}>    1.00
# 3 <{item=A},
#    {item=B}>    1.00
# 4 <{item=B},
#    {item=B}>    0.75
# 5 <{item=A},
#    {item=A}>    1.00
# 6 <{item=B},
#    {item=A}>    0.75

#lab4:mingap：所有的相邻的两个eventID的差大于mingap
#英文解释mingap: an integer value specifying the minimum time difference between consecutive elements of a sequence (default none, range >= 0).
s_result<-cspade(tmp_data.tran,parameter = list(support = 0.75,mingap=19),control = list(verbose = TRUE))
inspect(s_result)
#   items      support
# 1  <a href="http://4seohunt.biz/rep/bannerweb.upstate.edu">bannerweb.upstate.edu</a> <{item=A}>     1.00
# 2 <{item=B}>    1.00
# 3 <{item=A},
#    {item=B}>    1.00
# 4 <{item=A,
#     item=B}>    0.75
# 5 <{item=B},
#    {item=B}>    0.75
# 6 <{item=A,
#     item=B},
#    {item=B}>    0.75
#分析：缺少了实验lab1中的
# 7 <{item=A},
#    {item=A}>    1.00
#eventid只差分别是：sid=1时，eventid=30-10=20>19。sid=2时，eventid=30-20=10不>19。sid=3时，eventid=40-10=30>19。sid=4时，eventid=40-30=10不>19。即supp=2/4=0.5
# 8 <{item=B},
#    {item=A}>    0.75
#eventid只差分别是：sid=1时，eventid=30-10=20>19。sid=2时，无b->a。sid=3时，eventid=40-30=10不>19。sid=4时，eventid=40-30=10不>19。即supp=1/4=0.25

#lab5:maxgap：所有的相邻的两个eventID的差小于等于maxgap
#英文解释maxgap: an integer value specifying the maximum time difference between consecutive elements of a sequence (default none, range >= 0).
s_result<-cspade(tmp_data.tran,parameter = list(support = 0.75,maxgap=19),control = list(verbose = TRUE))
inspect(s_result)
#   items      support
# 1 <{item=A}>    1.00
# 2 <{item=B}>    1.00
# 3 <{item=A},
#    {item=B}>    0.75
# 4 <{item=A,
#     item=B}>    0.75
# 5 <{item=B},
#    {item=A}>    0.75
#分析：和实验lab1中数据的差别
#lab1中
# 3 <{item=A},
#    {item=B}>    1
#lab5中
# 3 <{item=A},
#    {item=B}>    0.75
#eventid只差分别是：sid=1时，eventid=30-20=10小于等于19，30-10=20不小于等于19。sid=2时，50-20=30不小于等于19,50-30=20不小于等于19,30-20=10小于等于19。sid=3时，eventid=30-10=20不小于等于19。sid=4时，eventid=40-30=10不小于等于19，50-40=10小于等于19。即supp=3/4=0.75