Skip to content

Commit d5d1a89

Browse files
committed
Added additional guidance on demo BDT EN NL
1 parent 136a90b commit d5d1a89

File tree

6 files changed

+80
-109
lines changed

6 files changed

+80
-109
lines changed

notebooks/CART_LawSchoolAdmissionBar.ipynb

Lines changed: 73 additions & 102 deletions
Large diffs are not rendered by default.

src/bias-detection.html

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@
33
<meta charset="UTF-8" />
44
<link rel="icon" href="/favicon.ico" type="image/x-icon" />
55
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
6-
<title>Algorithm Audit Bias Detection</title>
6+
<title>Unsupervised bias detection tool Algorithm Audit</title>
77
<meta http-equiv="CACHE-CONTROL" content="NO-CACHE" />
88
<meta http-equiv="PRAGMA" content="NO-CACHE" />
99
<link

src/index.html

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@
33
<meta charset="UTF-8" />
44
<link rel="icon" href="/favicon.ico" type="image/x-icon" />
55
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
6-
<title>Algorithm Audit Tools</title>
6+
<title>Open source AI auditing tools Algorithm Audit</title>
77
<meta http-equiv="CACHE-CONTROL" content="NO-CACHE" />
88
<meta http-equiv="PRAGMA" content="NO-CACHE" />
99
<link

src/locales/en.json

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -114,11 +114,11 @@
114114

115115
"biasAnalysis": {
116116
"demo": {
117-
"heading": "Information about demo dataset",
118-
"description": "A demo dataset is loaded below. Based on features of tweets, from the dataset [Twitter15], a BERT-based classifier has predicted whether a tweet is fake news or not. False positive classifications are marker in the FP column and will be used as a performance metric is this use case."
117+
"heading": "Information about demo dataset",
118+
"description": "As a demo, the [Twitter15](https://www.dropbox.com/scl/fi/flgahafqckxtup2s9eez8/rumdetect2017.zip?dl=0&e=1&file_subpath=%2Frumor_detection_acl2017%2Ftwitter15&rlkey=b7v86v3q1dpvcutxqk0xi7oej) dataset is loaded below. The dataset contains features of tweets and the prediction of a BERT-based misinformation classifier whether the tweet is fake news or not. False positive classifications are marked as FP. The FP column indicates that a tweet is falsy accused by the misinformation detection algorithm to be fake news. The FP metric will be used as a performance metric to measure bias is this use case.\n \n&nbsp;&nbsp;\n\n In this example, we will examine what type of tweets are more/less often classified by misinformation detection algorithm as fake news."
119119
},
120120
"parameters": {
121-
"heading": "Parameters selected",
121+
"heading": "Parameters selected for clustering",
122122
"iterations": "Number of iterations: {{value}}",
123123
"minClusterSize": "Minimal cluster size: {{value}}",
124124
"performanceMetric": "Performance metric column: {{value}}",

src/locales/nl.json

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -113,7 +113,7 @@
113113
"biasAnalysis": {
114114
"demo": {
115115
"heading": "Informatie over de demodataset",
116-
"description": "Een demodataset is hieronder geladen. Op basis van kenmerken van tweets, uit de dataset [Twitter15], heeft een op BERT-gebaseerde classifier voorspeld of een bericht nepnieuws is of niet. Fout-positieve classificaties zijn gemarkeerd in de FP-kolom en zullen in dit voorbeeld worden gebruikt als prestatiemetriek."
116+
"description": "Als demo wordt de [Twitter15](https://www.dropbox.com/scl/fi/flgahafqckxtup2s9eez8/rumdetect2017.zip?dl=0&e=1&file_subpath=%2Frumor_detection_acl2017%2Ftwitter15&rlkey=b7v86v3q1dpvcutxqk0xi7oej) dataset hieronder geladen. De dataset bevat kenmerken van tweets en de voorspelling van een BERT-gebaseerd misinformatie detectie algoritme of een tweet nepnieuws is of niet. Fout-positieve classificaties zijn gemarkeerd als FP. Een FP geeft aan dat tweet ten onrechte door het misinformatie algoritme zijn geclassificeerd als nepnieuws. De FP-metriek wordt in dit voorbeeld gebruikt als metriek om bias te meten. \n \n&nbsp;&nbsp;\n\n In dit voorbeeld onderzoeken we welk type tweets vaker/minder vaak door het misinformatie algoritme worden geclassificeerd als nepnieuws.""
117117
},
118118
"parameters": {
119119
"heading": "Geselecteerde parameters",

src/synthetic-data.html

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -3,7 +3,7 @@
33
<meta charset="UTF-8" />
44
<link rel="icon" href="/favicon.ico" type="image/x-icon" />
55
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
6-
<title>Algorithm Audit Synthetic Data Generaton</title>
6+
<title>Synthetic data generaton tool Algorithm Audit</title>
77
<meta http-equiv="CACHE-CONTROL" content="NO-CACHE" />
88
<meta http-equiv="PRAGMA" content="NO-CACHE" />
99
</head>

0 commit comments

Comments
 (0)