Last active
July 30, 2017 22:57
-
-
Save abicky/dfdeaec9efded25b733a to your computer and use it in GitHub Desktop.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
# Copyright 2015- Takeshi Arabiki | |
# License: MIT License (http://opensource.org/licenses/MIT) | |
library(data.table) | |
access_logs <- (function() { | |
set.seed(0) | |
USER_COUNT <- 10000 | |
PAGES <- c("page1", "page2", "page3") | |
STATES <- c("new", "normal", "active", "inactive") | |
INIT_PAGE_PROB <- structure(c(0.7, 0.2, 0.1), names = PAGES) | |
PAGE_TRANS_PROB <- matrix(c( | |
0.1, 0.4, 0.5, | |
0.8, 0.1, 0.1, | |
0.8, 0.1, 0.1 | |
), 3, byrow = TRUE, dimnames = list(PAGES, PAGES)) | |
STATE_TRANS_PROB <- matrix(c( | |
0.4, 0.2, 0.2, 0.2, | |
0.0, 0.4, 0.2, 0.4, | |
0.0, 0.1, 0.9, 0.0, | |
0.0, 0.1, 0.0, 0.9 | |
), 4, byrow = TRUE, dimnames = list(STATES, STATES)) | |
page_interval_time <- function(page) { | |
mean_interval <- switch(page, page1 = 30, 10) | |
return(as.integer(rnbinom(1, 1, mu = mean_interval - 1) + 1)) | |
} | |
session_interval_time <- function(state) { | |
interval_days <- switch( | |
state, | |
new = sample(1:3, 1, prob = c(0.7, 0.2, 0.1)), | |
normal = rpois(1, 2) + 1, | |
active = sample(1:3, 1, prob = c(0.7, 0.2, 0.1)), | |
inactive = sample(12, 1) * 30 | |
) | |
return(as.integer(interval_days * 86400 + rnorm(1, 0, 3600))) | |
} | |
access_count_per_session <- function() { | |
mean_access_count <- 3 | |
return(rpois(1, mean_access_count - 1) + 1) | |
} | |
HOUR_ACCESS_RATIO <- c(rep(0.1, 9), rep(1, 12), rep(0.1, 3)) | |
BASE_TIME <- as.integer(as.POSIXct("2015-05-03")) | |
MAX_TIME <- as.integer(as.POSIXct("2015-06-06")) | |
MAX_USER_ID <- 999999999 | |
BLOCK_SIZE <- 100000 | |
user_ids <- sample(MAX_USER_ID, USER_COUNT) | |
hours <- sample(24, USER_COUNT, replace = TRUE, prob = HOUR_ACCESS_RATIO) - 1L | |
first_access_times <- BASE_TIME + hours * 3600L + sample(3600, USER_COUNT, replace = TRUE) - 1L | |
log_times <- integer(BLOCK_SIZE) | |
log_user_ids <- integer(BLOCK_SIZE) | |
log_pages <- character(BLOCK_SIZE) | |
current_row <- 0L | |
for (i in seq_along(user_ids)) { | |
user_id <- user_ids[i] | |
time <- first_access_times[i] | |
state <- "new" | |
repeat { | |
page <- sample(PAGES, 1, prob = INIT_PAGE_PROB) | |
access_count <- access_count_per_session() | |
repeat { | |
current_row <- current_row + 1L | |
if (current_row > length(log_times)) { | |
log_times <- c(log_times, integer(length(log_times))) | |
log_user_ids <- c(log_user_ids, integer(length(log_user_ids))) | |
log_pages <- c(log_pages, character(length(log_pages))) | |
} | |
log_times[current_row] <- time | |
log_user_ids[current_row] <- user_id | |
log_pages[current_row] <- page | |
access_count <- access_count - 1 | |
if (access_count == 0) { | |
break | |
} | |
time <- time + page_interval_time(page) | |
page <- sample(PAGES, 1, prob = PAGE_TRANS_PROB[page, ]) | |
} | |
state <- sample(STATES, 1, prob = STATE_TRANS_PROB[state, ]) | |
time <- time + session_interval_time(state) | |
if (time >= MAX_TIME) { | |
break | |
} | |
} | |
} | |
access_logs <- data.table( | |
time = log_times[1:current_row], | |
user_id = log_user_ids[1:current_row], | |
page = log_pages[1:current_row] | |
) | |
access_logs[, c("idate", "itime") := IDateTime(as.POSIXlt(time, origin = "1970-01-01"))] | |
return(access_logs) | |
})() |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment