pars: Create or update a context
In thtrieu/edmsyn: Educational Data Synthesizer

Description Usage Arguments Details Value Author(s) See Also Examples

This function allows user assemble a new context or update an available context

pars(old.pars = NULL, default.vals = default(), dis = NULL, dif = NULL,
  abi = NULL, abi.mean = NULL, abi.sd = NULL, st.exp = NULL,
  it.exp = NULL, items = NULL, concepts = NULL, students = NULL,
  state = NULL, po = NULL, or.t = NULL, or.f = NULL,
  student.var = NULL, avg.success = NULL, min.ntree = NULL,
  max.ntree = NULL, trans = NULL, min.depth = NULL, max.depth = NULL,
  density = NULL, min.it.per.tree = NULL, max.it.per.tree = NULL,
  alpha.c = NULL, alpha.p = NULL, p.min = NULL, slip = NULL,
  guess = NULL, per.item = NULL, order = NULL, Q = NULL, S = NULL,
  M = NULL, L = NULL, bkt.mod = NULL, S.st.var = NULL,
  S.con.exp = NULL, L.st.var = NULL, L.con.exp = NULL,
  skill.space.size = NULL, skill.space = NULL, skill.dist = NULL,
  concept.exp = NULL, bkt.slip = NULL, bkt.guess = NULL, time = NULL,
  bkt.slip.it.exp = NULL, bkt.slip.st.var = NULL, bkt.guess.it.exp = NULL,
  bkt.guess.st.var = NULL, irt = NULL, exp = NULL, dina = NULL,
  dino = NULL, nmf.con = NULL, nmf.dis = NULL, nmf.com = NULL,
  lin.avg = NULL, poks = NULL, bkt = NULL, ...)

`old.pars`	an object of `context` class describe the context that needed to be updated, leave this parameter `NULL` if a new context is needed.
`default.vals`	an environment contains default values for some parameters in the context, by default it is initialized by function `default`
`dis`	a vector of discrimination values for each item
`dif`	a vector of difficulty values for each item
`abi`	a vector of ability values for each student
`abi.mean`	mean value of parameter `abi`
`abi.sd`	standard deviation of parameter `abi`
`st.exp`	a vector of expected success rates for each student
`it.exp`	a vector of expected success rates for each item
`items`	number of items
`concepts`	number of concepts
`students`	number of students
`state`	parameter for generating data from POKS model
`po`	dependency matrix of a partial order knowledge structure among items
`or.t`	parameter for generating data from POKS model
`or.f`	parameter for generating data from POKS model
`student.var`	variance of student expected success rate
`avg.success`	mean value of the response matrix
`min.ntree`	minimum number of connected components of `po`
`max.ntree`	maximum number of connected components of `po`
`trans`	a boolean value indicates if transitive links are allowed in `po`
`min.depth`	minimum depth of the connected components of `po`
`max.depth`	maximum depth of the connected components of `po`
`density`	a real value between 0 and 1, indicates the connection density of `po`
`min.it.per.tree`	minimum number of items per each connected component of `po`
`max.it.per.tree`	maxinum number of items per each connected component of `po`
`alpha.c`	parameter for learning by POKS model, see reference
`alpha.p`	parameter for learning by POKS model, see reference
`p.min`	p-value for interaction test while constructing POK structure
`slip`	a vector of slip factor for each item
`guess`	a vector of guess factor for each item
`per.item`	a boolean value indicates if the students can improve after taking each item
`order`	a vector indicates in which order did the students take the test in case `per.item` is set to be `TRUE`
`Q`	Q-matrix with size `items` times `concepts`
`S`	Skill matrix with size `concepts` times `students`
`M`	Skill mastery matrix with size`concepts` times `students`
`L`	Learn matrix indicates the transition probabilities for `M` matrix
`bkt.mod`	a character string indicates which model governs the generating process for sequential data
`S.st.var`	variance of student expected success rates of matrix `S`
`S.con.exp`	a vector of expected success rate for each concept in matrix `S`
`L.st.var`	variance of student expected success rates of matrix `L`
`L.con.exp`	a vector of expected success rate for each concept in matrix `L`
`skill.space.size`	size of the skill space
`skill.space`	a matrix with size `concepts` times `skill.space.size`
`skill.dist`	a vector of length `skill.space.size` that sums to one, indicates the probability of each skill pattern in `skill.space.size`
`concept.exp`	a vector of expected mastery rate for each concept
`bkt.slip`	a matrix of size `items` times `students` indicates slip factor for each combination of one item and one student
`bkt.guess`	a matrix of size `items` times `students` indicates slip factor for each combination of one item and one student
`time`	the number of time steps for sequential data
`bkt.slip.it.exp`	a vector of expected value for each item in `bkt.slip`
`bkt.slip.st.var`	variance of expected values of students in `bkt.slip`
`bkt.guess.it.exp`	a vector of expected value for each item in `bkt.guess`
`bkt.guess.st.var`	variance of expected values of students in `bkt.guess`
`irt`	a list with one component `R` being the response matrix, use in case of IRT model
`exp`	a list with one component `R` being the response matrix, use in case of expected model
`dina`	a list with two components `R` and `Q`, being the response matrix and Q matrix respectively, use in case of DINA model
`dino`	a list with two components `R` and `Q`, being the response matrix and Q matrix respectively, use in case of DINO model
`nmf.con`	a list with two components `R` and `concepts`, being the response matrix and number of concepts, use in case of NMF CONJUNCTIVE model
`nmf.dis`	a list with two components `R` and `concepts`, being the response matrix and number of concepts, use in case of NMF DISJUNCTIVE model
`nmf.com`	a list with two components `R` and `concepts`, being the response matrix and number of concepts, use in case of NMF COMPENSATORY model
`lin.avg`	a list with two components `R` and `concepts`, being the response matrix and number of concepts, use in case of LINEAR AVERAGE model
`poks`	a list with four components `R`, `alpha.p`, `alpha.c`, `p.min`, use in case of POKS model
`bkt`	a list with two components `R` and `order`, being the response matrix and its corresponding `order` vector (in case student improvement is allowed between taking two items), use in case of sequential data

This function takes in a set of parameters that the user input and assembles them into a context object, also checks for some simple types of potential conflicts

an object of context class describes the updated or newly assembled context

Hoang-Trieu Trinh, thtrieu@apcs.vn

init

# Declare a context where there are 15 students and 20 items
p <- pars(students = 15, items = 20)
class(p)
# See all parameters inside p
names(p)
# See the currently available parameters in p
print(p)