FIG. 9 lists feature tokens detected in handwriting and used to build n-gram feature vector histograms. In various embodiments, to classify the language, almost 100 individual features were used to capture the uniqueness of each language. For example, the French phrase, “Où dans la forêt le gar?on étudiant na?f?” illustrates all five French accent marks: 1) grave; 2) circumflex; 3) cedilla; 4) acute; and 5) umlaut. The appearance of these marks are detected and encoded as features. Each feature was assigned a number. Detectors were designated for each of features. There were other features such as a unique arrangement of circles and lines found in Korean, “![custom character]()
”, (“beauty is in the eye of the beholder”), the curves of Arabic, “![custom character]()
”, (“be patient”), the multiple orthogonal intersections of Chinese, “![custom character]()
” (“l(fā)ove at first sight”), and so on for Japanese, Urdu, Persian, Bengali, Hindu, Portuguese, Russian, Swahili, Tamil, Telugu, and Turkish.
