tokens
sequence
label_ids
sequence
[ "A", "SHAP", "summary", "plot", "shows", "the", "SHAP", "value", "of", "features", "for", "individual", "data", "points", "." ]
[ 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0 ]
[ "Figure", "~", "shows", "two", "examples", "of", "SHAP", "summary", "plots", "." ]
[ 0, 0, 0, 0, 0, 0, 1, 0, 0, 0 ]
[ "In", "each", "SHAP", "summary", "plot", "the", "horizontal", "axis", "indicates", "the", "SHAP", "value", ",", "the", "vertical", "axis", "indicates", "the", "features", ",", "and", "the", "feature", "values", "are", "shown", "using", "the", "color", "range", ",", "from", "blue", "(", "low", ")", "to", "red", "(", "high", ")", "." ]
[ 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "A", "positive", "SHAP", "value", "in", "our", "experiments", "indicates", "the", "impact", "of", "the", "feature", "towards", "the", "Benign", "class", "and", "a", "negative", "SHAP", "value", "indicates", "the", "impact", "towards", "the", "Attack", "class", "." ]
[ 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Each", "point", "in", "this", "plot", "is", "created", "by", "two", "values", ",", "the", "feature", "and", "the", "SHAP", "values", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Wherever", "multiple", "samples", "have", "the", "same", "SHAP", "value", ",", "their", "representative", "dot", "-", "points", "are", "piled", "up", "in", "a", "histogram", "manner", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "Since", "it", "is", "not", "possible", "to", "investigate", "the", "SHAP", "summary", "plots", "of", "all", "the", "experiment", "pairs", "separately", ",", "we", "chose", "three", "pairs", "of", "experiments", "with", "the", "maximum", "contrasting", "results", "that", "explain", "the", "main", "aspects", "of", "generalisability", "observed", "in", "the", "results", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "Hence", ",", "the", "larger", "height", "of", "the", "pile", "of", "dots", "indicates", "features", "of", "more", "sample", "points", "have", "the", "same", "SHAP", "value", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Since", "in", "these", "summary", "plots", ",", "only", "ten", "features", "are", "shown", ",", "a", "sample", "form", "a", "dataset", "corresponds", "with", "ten", "dots", "(", "data", "points", ")", "in", "a", "summary", "plot", ",", "one", "point", "per", "feature", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "The", "SHAP", "summary", "plot", "seems", "an", "appropriate", "tool", "for", "this", "kind", "of", "analysis", "." ]
[ 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "While", "having", "a", "high", "performance", "in", "a", "single", "domain", "evaluation", "is", "a", "necessary", "condition", "for", "any", "NIDS", "targeting", "the", "real", "-", "world", "application", ",", "there", "are", "other", "conditions", "to", "be", "met", "as", "well", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "One", "such", "condition", "is", "the", "generalisability", ",", "i.e.", ",", "the", "ability", "to", "generalise", "and", "translate", "the", "high", "performance", "on", "a", "single", "domain", "to", "other", "domains", "." ]
[ 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "In", "order", "to", "evaluate", "the", "generalisability", "of", "an", "NIDS", "model", ",", "we", "need", "to", "separate", "domains", "/", "datasets", "." ]
[ 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "The", "first", "domain", "/", "dataset", ",", "which", "is", "called", "the", "source", "domain", ",", "is", "used", "for", "the", "training", ",", "and", "the", "second", "domain", "/", "dataset", ",", "which", "is", "referred", "to", "as", "the", "target", "domain", ",", "is", "used", "for", "the", "evaluation", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0 ]
[ "In", "this", "way", ",", "the", "NIDS", "model", "is", "only", "trained", "on", "the", "source", "domain", "and", "it", "does", "not", "see", "the", "target", "domain", "during", "training", "." ]
[ 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0 ]
[ "Supervised", "Learning" ]
[ 0, 1 ]
[ "Here", "we", "evaluate", "the", "generalisability", "of", "the", "same", "four", "supervised", "learning", "models", "considered", "in", "the", "previous", "section", "." ]
[ 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0 ]
[ "We", "ran", "a", "set", "of", "four", "experiments", "for", "each", "NIDS", "model", "in", "which", "the", "model", "is", "trained", "on", "a", "source", "benchmark", "dataset", "and", "evaluated", "against", "the", "other", "three", "benchmark", "datasets", "without", "any", "further", "training", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Hence", ",", "for", "each", "target", "domain", "/", "dataset", "we", "have", "three", "different", "results", "for", "the", "same", "NIDS", "model", ",", "each", "indicating", "the", "performance", "of", "the", "model", "when", "trained", "on", "a", "different", "source", "domain", "/", "dataset", "." ]
[ 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0 ]
[ "Table", "~", "shows", "the", "results", "of", "these", "experiments", ",", "ordered", "by", "the", "target", "domain", "and", "source", "domain", ",", "to", "allow", "easy", "comparison", "with", "the", "single", "domain", "results", "shown", "in", "Table", "~", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "As", "can", "be", "seen", ",", "each", "supervised", "learning", "-", "based", "NIDS", "model", "is", "evaluated", "for", "the", "12", "different", "source", "/", "target", "domain", "combinations", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Figure", "~", "provides", "a", "more", "visual", "representation", "of", "both", "the", "single", "-", "domain", "and", "the", "multi-domain", "evaluation", "results", ",", "with", "each", "sub-figure", "showing", "the", "F1", "-", "Score", "results", "for", "a", "different", "supervised", "model", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0 ]
[ "The", "used", "colour", "map", "indicates", "a", "very", "high", "F1", "-", "Score", "(", "100\\%", ")", "in", "dark", "blue", ",", "and", "a", "very", "low", "value", "(", "0\\%", ")", "in", "light", "yellow", ",", "with", "the", "in", "-", "between", "values", "as", "indicated", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "The", "source", "datasets", "are", "indicated", "on", "the", "vertical", "axis", ",", "and", "the", "corresponding", "target", "datasets", "are", "shown", "on", "the", "horizontal", "axis", "." ]
[ 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0 ]
[ "On", "the", "diagonal", ",", "we", "observe", "the", "single", "-", "domain", "results", ",", "where", "a", "model", "is", "trained", "and", "evaluated", "on", "the", "same", "dataset", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "The", "off", "-", "diagonal", "results", "show", "the", "multi-domain", "results", ",", "which", "provide", "information", "about", "the", "degree", "of", "generalisability", "of", "the", "different", "models", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0 ]
[ "Machine", "learning", "(", "ML", ")", "based", "network", "intrusion", "detection", "systems", "(", "NIDSs", ")", "have", "been", "around", "for", "many", "years", "to", "address", "the", "shortcomings", "of", "signature", "-", "based", "NIDSs", "." ]
[ 0, 1, 0, 1, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "While", "a", "large", "number", "of", "methods", "have", "been", "proposed", "in", "the", "academic", "literature", "of", "NIDS", "with", "near", "perfect", "detection", "and", "classification", "performances", ",", "the", "ML", "-", "based", "NIDSs", "rarely", "have", "been", "used", "in", "the", "real", "-", "world", "scenarios", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "In", "these", "experiments", ",", "each", "ML", "-", "based", "NIDS", "is", "trained", "on", "a", "dataset", "and", "evaluated", "against", "all", "the", "four", "datasets", ",", "including", "the", "one", "used", "for", "its", "training", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "In", "this", "paper", "we", "extensively", "evaluate", "the", "generalisability", "of", "seven", "supervised", "and", "unsupervised", "ML", "-", "based", "NIDSs", "across", "four", "recently", "published", "publicly", "available", "NIDS", "datasets", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "As", "such", ",", "the", "lack", "of", "generalisation", "of", "the", "model", "performance", "from", "one", "domain", "to", "the", "other", ",", "and", "the", "asymmetric", "behaviour", "of", "models", "generalisability", "can", "all", "be", "attributed", "to", "the", "presence", "of", "these", "simple", "rule", "classifiers", "for", "one", "combination", "of", "model", "-", "source", "-", "target", "datasets", "and", "its", "absence", "in", "other", "combinations", "due", "to", "different", "feature", "distributions", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "The", "last", "observation", "indicates", "that", "the", "unsupervised", "ML", "-", "based", "NIDSs", "generalise", "better", "than", "the", "supervised", "ML", "-", "based", "NIDSs", ",", "even", "though", "their", "single", "domain", "performance", "is", "lower", "than", "the", "supervised", "ML", "-", "based", "NIDSs", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "We", "have", "further", "explained", "our", "results", "by", "finding", "the", "SHAP", "values", "for", "the", "model", "outputs", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0 ]
[ "Comparing", "the", "SHAP", "values", "of", "different", "dataset", "-", "model", "combinations", "indicates", "that", "the", "high", "classification", "performances", "in", "a", "combination", "of", "the", "model", "and", "source", "-", "target", "domains", "is", "mainly", "due", "to", "having", "one", "or", "more", "features", "that", "have", "strong", "correspondence", "with", "the", "Attack", "/", "Benign", "classes", "(", "positive", "and", "negative", "SHAP", "values", ")", "." ]
[ 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0 ]
[ "This", "makes", "it", "possible", "to", "compare", "a", "model", "'s", "performance", "in", "a", "single", "-", "domain", "and", "multi-domain", "(", "generalisability", ")", "evaluation", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 1, 0, 1, 0, 0, 0 ]
[ "The", "other", "observation", "is", "that", "generalisability", "can", "be", "asymmetric", ",", "which", "means", "performance", "of", "the", "model", "can", "significantly", "change", "when", "the", "source", "and", "target", "domains", "are", "swapped", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0 ]
[ "The", "results", "indicate", "that", "while", "some", "models", "are", "able", "to", "generalise", "over", "one", "or", "two", "datasets", ",", "none", "of", "the", "studied", "models", "generalise", "well", "across", "all", "datasets", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0 ]
[ "Then", "we", "compare", "the", "results", "of", "single", "domain", "evaluation", ",", "in", "which", "the", "same", "benchmark", "dataset", "is", "used", "for", "both", "training", "and", "evaluation", ",", "with", "the", "results", "of", "generalisability", "evaluation", "in", "which", "a", "trained", "model", "is", "evaluated", "against", "the", "other", "three", "benchmark", "datasets", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "Finally", ",", "we", "apply", "the", "Shapley", "values", "~", "on", "data", "and", "models", "in", "different", "scenarios", "and", "explain", "the", "generalisability", "by", "comparing", "the", "Shapley", "values", "for", "different", "experiments", "." ]
[ 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0 ]
[ "While", "there", "is", "no", "previous", "work", "systematically", "evaluating", "the", "generalisability", "of", "ML", "-", "based", "NIDSs", ",", "there", "are", "two", "previous", "studies", "that", "partially", "investigate", "evaluation", "of", "ML", "-", "based", "NIDSs", "via", "datasets", "unseen", "during", "training", "which", "include", "~", "and", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0 ]
[ "In", "~", "it", "is", "shown", "that", "the", "near", "perfect", "performance", "of", "simple", "supervised", "ML", "models", "on", "the", "dataset", "used", "for", "their", "training", "considerably", "drops", "when", "evaluated", "on", "other", ",", "unseen", "benchmark", "datasets", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "This", "study", "uses", "the", "Kyoto", "+", "~", ",", "gure", "KDD", "~", "and", "NSL", "-", "KDD~", "NIDS", "datasets", "which", "are", "published", "in", "2006", ",", "2008", "and", "2009", "respectively", "." ]
[ 0, 0, 0, 0, 1, 1, 0, 0, 0, 1, 0, 0, 1, 1, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "However", ",", "the", "Kyoto", "+", "dataset", "is", "not", "included", "in", "the", "evaluation", "of", "unseen", "datasets", ",", "due", "to", "the", "difference", "in", "the", "feature", "set", "with", "other", "datasets", "." ]
[ 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "While", "this", "study", "discusses", "cross", "evaluation", "of", "ML", "-", "based", "NIDSs", ",", "it", "does", "not", "investigate", "the", "factors", "for", "the", "poor", "performance", "on", "the", "ML", "-", "models", "on", "unseen", "benchmark", "datasets", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0 ]
[ "In", "addition", ",", "the", "results", "presented", "in", "~", "are", "based", "on", "only", "two", "old", "benchmark", "datasets", ",", "which", "makes", "it", "hard", "to", "draw", "conclusions", "about", "the", "generalisability", "of", "ML", "-", "based", "NIDSs", "more", "broadly", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0 ]
[ "Furthermore", ",", "the", "study", "only", "includes", "the", "supervised", "learning", "models", "and", "does", "not", "include", "an", "evaluation", "of", "unsupervised", "learning", "models", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "The", "~", "is", "the", "other", "work", "in", "which", "the", "cross", "evaluation", "of", "ML", "-", "based", "NIDSs", "is", "discussed", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0 ]
[ "The", "main", "focus", "of", "this", "study", "is", "proposing", "a", "framework", "for", "cross", "evaluation", "of", "ML", "-", "based", "NIDSs", "by", "creating", "various", "combinations", "of", "the", "Benign", "and", "Attack", "classes", "of", "one", "or", "more", "datasets", "for", "the", "training", "and", "evaluation", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0 ]
[ "The", "proposed", "framework", "consists", "of", "10", "different", "combinations", "in", "which", "combination", "number", "4", "considers", "generalisation", "capabilities", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "In", "this", "scenario", "while", "the", "Benign", "class", "of", "the", "same", "dataset", "is", "used", "for", "the", "model", "training", "and", "evaluation", ",", "the", "attack", "classes", "of", "different", "datasets", "are", "used", "for", "the", "training", "and", "evaluation", "." ]
[ 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0 ]
[ "Then", "they", "provide", "the", "results", "for", "this", "generalisation", "evaluation", ",", "and", "show", "that", "there", "is", "a", "significant", "performance", "drop", "in", "detecting", "various", "attacks", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "The", "study", "uses", "the", "same", "4", "datasets", "used", "in", "our", "study", "for", "the", "uniform", "scenarios", "but", "provides", "the", "detection", "ratios", "for", "the", "Botnet", ",", "DoS", "and", "Other", "(", "all", "the", "other", "classes", ")", "separately", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "While", "this", "is", "the", "closest", "study", "to", "what", "we", "are", "doing", "in", "this", "paper", ",", "using", "the", "Benign", "class", "of", "the", "same", "dataset", "in", "the", "training", "and", "evaluation", "makes", "it", "different", "from", "our", "work", "that", "evaluates", "generalisability", "over", "a", "full", "new", "dataset", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0 ]
[ "While", "this", "is", "the", "most", "prominent", "example", ",", "we", "observe", "a", "generally", "high", "degree", "of", "asymmetry", "of", "the", "generalisability", "across", "different", "source", "/", "target", "domain", "pairs", "and", "supervised", "learning", "models", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "This", "is", "consistent", "across", "all", "four", "ML", "models", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "If", "we", "swap", "the", "source", "and", "target", "domain", ",", "and", "use", "BoT", "-", "IoT", "as", "the", "source", "and", "UNSW", "-", "NB15", "as", "the", "target", ",", "the", "Extra", "Tree", "only", "achieves", "4.90\\%", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "For", "example", ",", "for", "the", "Extra", "Tree", "model", "(", "Figure", "~-", "a", ")", ",", "UNSW", "-", "NB15", "as", "the", "source", "domain", "generalises", "well", "to", "the", "BoT", "-", "IoT", "dataset", "as", "the", "target", "domain", ",", "with", "an", "F1", "-", "Score", "of", "94.83\\%", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "If", "we", "compare", "the", "generalisability", "results", "across", "the", "four", "different", "ML", "models", ",", "we", "observe", "some", "consistent", "patterns", ",", "but", "we", "also", "notice", "some", "significant", "differences", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "For", "instance", ",", "the", "cell", "in", "the", "first", "row", "and", "first", "column", "shows", "the", "average", "decay", "of", "the", "performance", "of", "the", "Extra", "Tree", "model", "trained", "on", "NFv2", "-", "BoT", "-IoT", "when", "tested", "on", "NFv2", "-", "CIC", "-", "2018", ",", "NFv2", "-", "ToN", "-IoT", "and", "NFv2", "-", "UNSW", "-", "NB15", "datasets", ",", "compared", "to", "its", "performance", "when", "tested", "on", "NFv2", "-", "BoT", "-", "IoT", "dataset", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Each", "column", "indicates", "a", "supervised", "learning", "model", "and", "each", "row", "indicates", "a", "source", "/", "training", "dataset", "." ]
[ 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "The", "mostly", "dark", "colouring", "on", "the", "diagonal", "indicates", "a", "generally", "high", "performance", "in", "the", "single", "-", "domain", "evaluation", "of", "all", "the", "four", "supervised", "models", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0 ]
[ "Most", "strikingly", ",", "we", "see", "that", "if", "the", "UNSW", "-", "NB15", "dataset", "is", "chosen", "as", "the", "target", "domain", ",", "the", "results", "are", "very", "poor", "for", "any", "of", "the", "other", "datasets", "chosen", "as", "the", "source", "domain", "(", "rightmost", "column", ")", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0 ]
[ "The", "Random", "Forest", "model", "seems", "to", "perform", "quite", "differently", "from", "the", "other", "three", "modeles", ",", "in", "particluar", "for", "the", "UNSW", "-", "NB15", "dataset", "as", "the", "source", "domain", "(", "bottom", "row", ")", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 0 ]
[ "However", ",", "the", "mostly", "lighter", "colours", ",", "and", "hence", "lower", "F1", "-", "Scores", ",", "indicate", "a", "generally", "poor", "ability", "of", "the", "models", "to", "generalise", "from", "a", "source", "dataset", "to", "a", "different", "target", "dataset", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "As", "can", "be", "seen", ",", "in", "both", "SHAP", "summary", "plots", "most", "of", "the", "features", "have", "a", "zero", "SHAP", "value", ",", "indicating", "no", "impact", "on", "model", "output", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "Though", ",", "in", "Figure", "~-", "(", "a", ")", "the", "low", "(", "blue", ")", "and", "high", "(", "red", ")", "values", "of", "the", "first", "important", "feature", ",", "MAX\\_TTL", ",", "have", "a", "clear", "correspondence", "with", "the", "Benign", "and", "Attack", "classes", "(", "low", "feature", "values", "have", "positive", "and", "high", "feature", "values", "have", "negative", "SHAP", "values", ")", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0 ]
[ "In", "Figure", "~-", "(", "a", ")", "the", "Extra", "Tree", "model", "is", "trained", "on", "NFv2", "-", "BoT", "-", "IoT", "dataset", "and", "evaluated", "on", "NFv2", "-", "ToN", "-", "IoT", "dataset", ",", "and", "in", "(", "b", ")", "it", "is", "trained", "on", "NFv2", "-", "ToN", "-", "IoT", "dataset", "and", "evaluated", "on", "NFv2", "-", "BoT", "-", "IoT", "dataset", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Finally", ",", "we", "investigate", "another", "case", "of", "asymmetric", "generalisability", "with", "a", "different", "distribution", "of", "SHAP", "values", "as", "shown", "in", "Figure", "~", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0 ]
[ "As", "can", "be", "seen", "in", "Figure", "~-", "(", "a", ")", "the", "SHAP", "values", "of", "all", "features", "except", "the", "MAX\\_TTL", "is", "zero", ",", "i.e.", "they", "do", "not", "affect", "either", "of", "classes", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "Hence", ",", "while", "a", "high", "performance", "is", "expected", "for", "the", "observed", "single", "-", "rule", "classifier", "in", "Figure", "~-", "(", "a", ")", "(", "with", "a", "huge", "difference", "between", "the", "SHAP", "values", "of", "the", "low", "and", "high", "MAX\\_TTL", "values", ")", ",", "it", "is", "hard", "to", "imagine", "a", "similar", "performance", "for", "the", "model", "in", "Figure", "~-", "(", "b", ")", "that", "has", "a", "single", "feature", "mostly", "indicating", "to", "the", "Benign", "class", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "The", "dataset", "samples", "with", "a", "low", "MAX", "\\_", "TTL", "value", "are", "classified", "as", "Benign", "and", "samples", "with", "a", "high", "MAX", "\\_", "TTL", "value", "are", "classified", "as", "Attack", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "A", "mix", "of", "low", "and", "high", "feature", "values", "can", "be", "seen", "in", "the", "positive", "and", "negative", "ranges", "of", "the", "SHAP", "value", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "However", ",", "in", "Figure", "~-", "(", "b", ")", "most", "of", "the", "MAX\\_TTL", "values", "have", "positive", "SHAP", "values", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Figure", "~", "shows", "the", "mean", "absolute", "SHAP", "value", "for", "the", "top", "ten", "features", "where", "the", "Feed", "forward", "model", "is", "trained", "on", "NFv2", "-", "BoT", "-", "IoT", "and", "evaluated", "on", "(", "a", ")", "NFv2-", "BoT", "-", "IoT", "and", "(", "b", ")", "NFv2-", "UNSW", "-", "NB15", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1, 1, 0 ]
[ "The", "features", "are", "sorted", "in", "terms", "of", "their", "overall", "average", "in", "descending", "order", ",", "and", "the", "overall", "average", "of", "the", "mean", "SHAP", "value", "of", "each", "feature", "is", "also", "shown", "by", "a", "red", "circle", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "Figure", "~", "shows", "the", "distribution", "of", "these", "mean", "SHAP", "values", "for", "the", "three", "unsupervised", "learning", "models", ",", "the", "Isolation", "Forest", ",", "oSVM", "and", "SGD", "-", "oSVM", "respectively", "plotted", "in", "Figure", "~", "-", "(", "a", ")", ",", "(", "b", ")", ",", "and", "(", "c", ")", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "For", "this", "purpose", ",", "we", "computed", "the", "mean", "SHAP", "value", "of", "all", "features", "across", "all", "the", "experiments", ",", "and", "plotted", "their", "distributions", "side", "by", "side", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "The", "horizontal", "axis", "indicates", "the", "features", "and", "vertical", "axis", "indicates", "the", "normalized", "mean", "SHAP", "value", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Since", "the", "range", "of", "mean", "SHAP", "values", "for", "different", "features", "were", "different", ",", "they", "have", "been", "normalized", "to", "make", "them", "comparable", "." ]
[ 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "For", "instance", ",", "in", "Figure", "~-", "(", "a", ")", ",", "PROTOCOL", "is", "the", "most", "important", "feature", "with", "a", "mean", "SHAP", "value", "of", "0.005", ",", "while", "in", "Figure", "~-", "(", "b", ")", ",", "the", "most", "important", "feature", "is", "FLOW", "\\_DURATION", "\\_IN", "\\_MILLISECONDS", "with", "a", "mean", "SHAP", "value", "of", "0.014", ",", "and", "PROTOCOL", "is", "fifth", "important", "feature", "with", "a", "mean", "SHAP", "value", "of", "0.008", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0 ]
[ "Comparing", "the", "two", "feature", "importance", "plots", "clearly", "shows", "that", "feature", "orders", "and", "the", "mean", "SHAP", "values", "of", "features", "significantly", "varies", "between", "the", "two", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "These", "figures", ",", "called", "the", "feature", "importance", "plot", ",", "show", "the", "features", "in", "descending", "order", "(", "of", "their", "mean", "of", "absolute", "SHAP", "values", ")", "on", "the", "vertical", "axis", "and", "the", "mean", "SHAP", "value", "on", "the", "horizontal", "axis", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0 ]
[ "This", "is", "a", "clear", "indication", "that", "the", "behaviour", "of", "the", "Feed", "Forward", "model", "trained", "on", "NFv2", "-", "BoT", "-", "IoT", "is", "entirely", "different", "when", "tested", "on", "NFv2", "-", "BoT", "-", "IoT", "datasets", "and", "when", "tested", "on", "NFv2", "-", "UNSW", "-", "NB15", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0 ]
[ "Accordingly", ",", "comparing", "the", "feature", "importance", "across", "all", "the", "16", "combinations", "of", "source", "-", "target", "domains", "will", "reveal", "the", "overall", "model", "behaviour", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0 ]
[ "This", "conclusion", "is", "consistent", "with", "the", "results", "shown", "in", "Figure", "~-", "(", "C", ")", "for", "the", "Feed", "Forward", "model", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Since", "the", "sizes", "of", "these", "datasets", "are", "different", ",", "e.g.", "NFv2", "-", "UNSW", "-", "NB15", "has", "2,390,275", "flows", "and", "the", "BoT", "-", "IoT", "dataset", "has", "37,763,497", "flows", ",", "we", "used", "a", "stratified", "(", "with", "regards", "to", "classes", ")", "sampling", "strategy", "with", "a", "size", "of", "1,000,000", "flows", "and", "we", "ran", "all", "the", "experiments", "on", "the", "sampled", "datasets", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "As", "can", "be", "seen", ",", "all", "these", "datasets", "come", "with", "a", "Benign", "class", "and", "various", "numbers", "of", "different", "attack", "classes", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Since", "the", "set", "of", "attack", "classes", "is", "different", "for", "each", "dataset", ",", "we", "therefore", "focus", "on", "binary", "classification", "only", ",", "i.e.", "Benign", "vs", "Attack", "traffic", "in", "our", "analysis", "." ]
[ 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0 ]
[ "Consequently", ",", "all", "attack", "classes", "of", "each", "dataset", "are", "aggregated", "under", "a", "single", "class", "called", "Attack", "." ]
[ 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Hence", ",", "the", "datasets", "used", "in", "this", "study", "have", "two", "classes", ",", "i.e.", "Benign", "and", "Attack", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 1, 0 ]
[ "Initially", ",", "when", "we", "ran", "the", "experiments", "related", "to", "unsupervised", "learning", "algorithms", ",", "we", "noticed", "that", "the", "performance", "(", "F1", "-", "Score", ")", "of", "the", "models", "was", "very", "poor", "when", "trained", "or", "evaluated", "on", "two", "datasets", "NFv2", "-", "BoT", "-", "IoT", "and", "NFv2", "-", "ToN", "-", "IoT", ",", "in", "both", "single", "domain", "or", "multi-domain", "evaluations", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "This", "was", "mainly", "due", "to", "the", "unrealistically", "high", "imbalance", "of", "the", "datasets", ",", "i.e.", "the", "ratio", "of", "the", "Attack", "to", "Benign", "records", ",", "as", "can", "be", "seen", "in", "Table", "~", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "Accordingly", ",", "we", "created", "and", "used", "a", "balanced", "version", "of", "all", "four", "datasets", "(", "via", "down", "sampling", ")", "in", "terms", "of", "Attack", "-", "Benign", "labels", ",", "and", "used", "for", "the", "unsupervised", "learning", "models", "experiments", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "These", "datasets", ",", "extended", "with", "a", "``", "-", "b", "\"", "suffix", "such", "as", "NFv2", "-", "BoT", "-", "IoT", "-", "b", ",", "indicate", "the", "balanced", "version", "of", "the", "(", "e.g.", "NFv2", "-", "BoT", "-", "IoT", ")", "dataset", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Single", "Domain", "Evaluation", "%" ]
[ 0, 0, 1, 0 ]
[ "In", "this", "section", "we", "evaluate", "the", "ML", "-", "based", "NIDSs", "using", "the", "common", "method", "of", "NIDS", "evaluation", "i.e.", ",", "using", "a", "single", "dataset", "for", "the", "training", "and", "evaluation", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0 ]
[ "In", "this", "method", ",", "a", "publicly", "available", "benchmark", "NIDS", "dataset", "is", "divided", "into", "the", "training", "and", "test", "subsets", ",", "and", "the", "ML", "model", "is", "trained", "and", "tested", "against", "these", "subsets", "of", "the", "same", "dataset", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
[ "However", ",", "the", "ML", "model", "is", "trained", "and", "tested", "/", "evaluated", "against", "each", "dataset", "separately", ",", "and", "the", "generalisability", "across", "datasets", "is", "not", "considered", "." ]
[ 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0 ]
[ "Since", "the", "training", "and", "test", "data", "are", "both", "collected", "from", "the", "same", "environment", ",", "this", "approach", "is", "referred", "to", "as", "single", "domain", "evaluation", "." ]
[ 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 0 ]
[ "Although", "the", "two", "above", "works", "partially", "study", "the", "generalisability", "of", "ML", "-", "based", "NIDSs", ",", "their", "main", "focus", "is", "a", "different", "subject", "." ]
[ 0, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0 ]
README.md exists but content is empty. Use the Edit dataset card button to edit it.
Downloads last month
0
Edit dataset card