diss_multifeature.png
This image displays a data table titled "Table 31: Accuracy by feature grouping in English (CVC/NVN) and French". The table is organized into eight columns and nine rows, including the header row. The columns represent different metrics for accuracy across two languages, English and French, using two different machine learning methods, SVM and RF.

The column headers are as follows:
1. Grouping
2. N (representing the number of items or samples)
3. en.svm.acc (English Support Vector Machine accuracy)
4. en.rf.acc (English Random Forest accuracy)
5. fr.svm.acc (French Support Vector Machine accuracy)
6. fr.rf.acc (French Random Forest accuracy)
7. avg.svm (Average SVM accuracy)
8. avg.rf (Average RF accuracy)

The rows list different feature groupings with their corresponding data values:

Row 1: All-Inclusive
- N: 29
- en.svm.acc: 84.764
- en.rf.acc: 84.539
- fr.svm.acc: 93.729
- fr.rf.acc: 91.566
- avg.svm: 89.247
- avg.rf: 88.052

Row 2: LowRedundancy
- N: 14
- en.svm.acc: 83.171
- en.rf.acc: 84.539
- fr.svm.acc: 91.896
- fr.rf.acc: 91.309
- avg.svm: 87.533
- avg.rf: 87.924

Row 3: Preliminary
- N: 10
- en.svm.acc: 82.290
- en.rf.acc: 83.902
- fr.svm.acc: 91.713
- fr.rf.acc: 90.906
- avg.svm: 87.001
- avg.rf: 87.404

Row 4: Importance10
- N: 10
- en.svm.acc: 82.084
- en.rf.acc: 83.883
- fr.svm.acc: 91.749
- fr.rf.acc: 90.942
- avg.svm: 86.917
- avg.rf: 87.413

Row 5: SingleFeature10
- N: 10
- en.svm.acc: 80.922
- en.rf.acc: 82.046
- fr.svm.acc: 88.009
- fr.rf.acc: 88.045
- avg.svm: 84.465
- avg.rf: 85.046

Row 6: ADW3
- N: 6
- en.svm.acc: 75.787
- en.rf.acc: 78.017
- fr.svm.acc: 84.745
- fr.rf.acc: 86.432
- avg.svm: 80.266
- avg.rf: 82.225

Row 7: A1-P0 Only
- N: 1
- en.svm.acc: 67.279
- en.rf.acc: 61.863
- fr.svm.acc: 76.128
- fr.rf.acc: 66.923
- avg.svm: 71.703
- avg.rf: 64.393

Row 8: Duration Only
- N: 1
- en.svm.acc: 65.640
- en.rf.acc: 79.171
- fr.svm.acc: 65.640
- fr.rf.acc: 79.171
- avg.svm: 65.640
- avg.rf: 79.171
This description was generated automatically. Please feel free to ask questions if you have further questions about the nature of the image or its meaning within the presentation.