커뮤니티에서 많은 텍스트를 다운로드했고 약 4개의 주제를 찾았습니다. 클러스터 2의 이름을 "JSL"로 지정하고 클러스터 3의 이름을 "그래프"로 지정할 수 있습니다.클러스터 4는 "DOE"이고 클러스터 1은 "통계"라고 생각합니다.
TE 사용법을 아는 누군가가 "멋져요. 그냥 하세요. 이것 모든 사용자 질문을 받아 커뮤니티에서 가장 가능성 있는 답변을 제시할 수 있는 모델을 갖게 될 것입니다."예, 저는 답변을 찾고 싶지 않은 또 다른 사용자입니다.
첨부된 데이터; 아마도 JMP 사람들은 여기 외부에서 공유하지 않기를 원할 것입니다. @Jeff_Perkinson
Text Explorer(
Text Columns( :description ),
Set Regex(
Library( "HTML Tag Remover" ),
Library( "HTML Escape Remover" ),
Library( "Words" )
),
Add Stop Words(
{"color", "coloration", "colored", "colores", "colorful", "coloring",
"colorings", "colorization", "colorize", "colors", "column", "column's",
"columne", "columns", "columns's", "courier", "data", "data's", "datas",
"familier", "families", "family", "font", "fonts", "jmp", "jmp's", "like",
"liked", "likeing", "likely", "likes", "liking", "model", "model's",
"modeled", "modeleing", "modeler", "modelers", "modeling", "modelization",
"modelizing", "modell", "modelled", "modelling", "modells", "models", "name",
"name's", "named", "namely", "names", "naming", "new", "newness", "one",
"one's", "ones", "row", "row's", "rowed", "rows", "script", "script's",
"scripte", "scripted", "scripting", "scripting's", "scriptly", "scripts",
"scriptting", "set", "set's", "sets", "setted", "setting", "setting's",
"settings", "size", "sized", "sizees", "sizes", "sizing", "style", "styled",
"styles", "styling", "tabl", "table", "table's", "tabled", "tables",
"tables's", "tabls", "thank", "thank's", "thankful", "thankfully",
"thanking", "thanks", "use", "used", "useful", "usefully", "usefulness",
"useing", "uses", "using", "valu", "value", "value's", "valued", "values",
"valuing", "work", "worked", "working", "workings", "works"}
),
Add Phrases(
{"access or evaluation", "add factor", "add factor continuous",
"add ref line", "add term add term", "alias term", "analysis of variance",
"analyze fit", "application builder", "associative array",
"associative arrays", "automatic recalc", "background map", "bar chart",
"best formula", "bivar plot", "books scripting guide", "bootstrap forest",
"box plot", "bubble plot", "button box", "capability analysis",
"categorical factors", "center points", "chart builder", "check box",
"chi square", "click the red triangle", "col mean", "col quantile",
"col sum", "combo box", "confidence interval", "continuous distribution",
"continuous factor", "continuous factors", "continuous formula",
"contour plot", "control chart", "control chart builder", "control charts",
"control limits", "control panel", "control panel variables", "csv files",
"custom design", "database connection", "date time", "definitive screening",
"degrees of freedom", "dependent variable", "design matrix",
"design platform", "dev chart", "dialog box", "dispatch scalebox",
"display box", "distribution platform", "double click", "drop zone",
"effect leverage", "elements position", "error bars", "eval parse",
"excel file", "experimental design", "factor continuous", "factor levels",
"factorial design", "files in directory", "fit curve", "fit least",
"fit least squares", "fit line", "fit model dialog", "fit model platform",
"fit spline", "formula editor", "formula random", "fractional factorial",
"framebox frame", "full factorial", "generalized linear", "goodness of fit",
"graph builder", "grouping variable", "help books", "inc minor ticks",
"independent variables", "individual measurement", "interaction effects",
"interactive html", "inverse prediction", "jsl code", "ksigma chart",
"lack of fit", "least squares", "least squares emphasis",
"life distribution", "linear regression", "list box", "local data filter",
"log window", "logistic regression", "main effect", "main effects",
"matched pairs", "measurement system", "min max inc", "minor ticks",
"mixture design", "moving average", "multiple comparisons",
"multivariate methods", "neural network", "nominal format",
"nominal logistic", "non linear", "nonlinear platform",
"normally distributed", "null hypothesis", "number col", "number col box",
"number of runs", "numeric continuous", "odbc driver", "odds ratio",
"overlay plot", "parameter estimate", "parameter estimates", "parse eval",
"pass fail", "pca fullmatrix", "personality standard",
"personality standard least", "personality standard least squares",
"pick file", "predicted plot", "prediction formula", "prediction profiler",
"principal components", "process capability", "process screening",
"properties line", "quality and process", "quantile plot", "query builder",
"radio box", "random effect", "random effects", "random seed",
"red triangle", "ref line", "reference line", "reference lines",
"regression analysis", "repeated measures", "right click", "run program",
"sample data semiconductor", "scatter plot", "screening platform",
"semiconductor capability", "show control panel", "spec limit",
"spec limits", "sql query", "sql statement", "standard deviation",
"standard deviations", "standard least", "standard least squares",
"standard least squares emphasis", "statistic mean", "std dev",
"std dev chart", "sum of squares", "summary statistic",
"summary statistic mean", "term add", "term add term", "text box",
"text file", "time points", "time series", "tree structure",
"variability chart", "variance components", "whole plot"}
),
Maximum Characters per Word( 20 ),
Minimum Characters per Word( 6 ),
Latent Class Analysis(
Number of Clusters(5 ),
Maximum Number of Terms( 7883 ),
Minimum Term Frequency( 10 ),
Set Random Seed( -2147483648 )
),
Stemming( "Stem for Combining" ),
Customize Regex( 0 ),
Language( "English" ),
SendToReport(
Dispatch(
{"Latent Class Analysis for 8 Clusters", "MDS Plot"},
"1",
ScaleBox,
{Min( -129.310134934137 ), Max( 121.950357086719 ), Inc( 50 ),
Minor Ticks( 0 )}
)
)
);
