Cognitive telephone fraud detection
Embodiments of the present invention provide a method, system and computer program product for telephonic fraud prevention. In an embodiment of the invention, a method of telephonic fraud prevention includes speech recognizing in real-time, audio of a telephone call from a caller to a called party. Once the real-time audio has been speech recognized into text, the text may be parsed into words and the words compared to a hierarchical semantic tree of terms. On the condition that a match is detected between one of the words and a node of the tree, a level of the tree is identified at which the match is detected and the level is correlated to a degree of risk of the telephone call. Finally, an alert is displayed to the called party reflecting the degree of risk.
1. A method for telephonic fraud prevention comprising:
speech recognizing in real-time, audio of a telephone call from a caller to a called party;
parsing text from the speech recognized audio and comparing words of the text to a hierarchical semantic tree of terms;
establishing different risk degrees for each level of a multiplicity of nodes of the hierarchical semantic tree of terms; and,
on condition that a match is detected between one of the words and a word of a node of the tree, identifying a furthest depth of the tree at which the one of the words and the term of the node match, identifying a level for the furthest depth of the tree, correlating the identified level to a degree of risk of the telephone call, and displaying an alert to the called party reflecting the degree of risk.
2. The method of claim 1 , further comprising:
submitting the speech recognized audio to a sentiment analysis engine to receive an indication of sentiment corresponding to the telephone call;
comparing the indication of sentiment to a pre-determined common sentiment for a classification of the telephone call; and,
performing the displaying only on upon the indication of sentiment differing from the common sentiment.
3. The method of claim 1 , further comprising:
submitting the speech recognized audio to a sentiment analysis engine to receive an indication of sentiment corresponding to the telephone call;
comparing the indication of sentiment to a pre-determined common sentiment for a classification of the telephone call; and,
additionally displaying a confirmation of the alert upon the indication of sentiment differing from the common sentiment.
4. The method of claim 1 , further comprising:
detecting a source phone number for the call and identifying a geographic location from which the telephone call originates;
determining if the geographic location is associated with a higher risk of a fraudulent telephone call; and,
performing the speech recognizing, parsing, identifying and displaying in response to determining that the geographic location is associated with a higher risk of a fraudulent telephone call.
5. The method of claim 2 , further comprising:
recording a set of telephone calls received by the called party;
classifying each telephone call in the set; and,
submitting each of the telephone calls in the set to a sentiment analysis engine in order to compute a sentiment common to each classification of the set.
6. The method of claim 3 , further comprising:
recording a set of telephone calls received by the called party;
classifying each telephone call in the set; and,
submitting each of the telephone calls in the set to a sentiment analysis engine in order to compute a sentiment common to each classification of the set.
7. A data processing system configured for telephonic fraud prevention, the system comprising:
a host computing system comprising memory and at least one processor and coupled to a telephone;
fixed storage coupled to the host computing system;
a hierarchical semantic tree of terms stored in the fixed storage; and,
a telephonic fraud prevention module comprising computer program instructions executing in the memory of the host computing system that upon execution are adapted to perform:
speech recognizing in real-time, audio of a telephone call processed in the telephone from a caller to a called party;
parsing text from the speech recognized audio and comparing words of the text to the hierarchical semantic tree of terms stored in the fixed storage;
establishing different risk degrees for each level of a multiplicity of nodes of the hierarchical semantic tree of terms; and,
on condition that a match is detected between one of the words and a word of a node of the tree, identifying a furthest depth of the tree at which the one of the words and the term of the node match, identifying a level for the furthest depth of the tree, correlating the identified level to a degree of risk of the telephone call, and displaying an alert to the called party reflecting the degree of risk.
8. The system of claim 7 , wherein the program instructions during execution are further adapted to perform:
submitting the speech recognized audio to a sentiment analysis engine to receive an indication of sentiment corresponding to the telephone call;
comparing the indication of sentiment to a pre-determined common sentiment for a classification of the telephone call; and,
performing the displaying only on upon the indication of sentiment differing from the common sentiment.
9. The system of claim 7 , wherein the program instructions during execution are further adapted to perform:
submitting the speech recognized audio to a sentiment analysis engine to receive an indication of sentiment corresponding to the telephone call;
comparing the indication of sentiment to a pre-determined common sentiment for a classification of the telephone call; and,
additionally displaying a confirmation of the alert upon the indication of sentiment differing from the common sentiment.
10. The system of claim 7 , wherein the program instructions during execution are further adapted to perform:
detecting a source phone number for the call and identifying a geographic location from which the telephone call originates;
determining if the geographic location is associated with a higher risk of a fraudulent telephone call; and,
performing the speech recognizing, parsing, identifying and displaying in response to determining that the geographic location is associated with a higher risk of a fraudulent telephone call.
11. The system of claim 8 , wherein the program instructions during execution are further adapted to perform:
recording a set of telephone calls received by the called party;
classifying each telephone call in the set; and,
submitting each of the telephone calls in the set to a sentiment analysis engine in order to compute a sentiment common to each classification of the set.
12. The system of claim 9 , wherein the program instructions during execution are further adapted to perform:
recording a set of telephone calls received by the called party;
classifying each telephone call in the set; and,
submitting each of the telephone calls in the set to a sentiment analysis engine in order to compute a sentiment common to each classification of the set.
13. A computer program product for telephonic fraud prevention, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, the program instructions executable by a device to cause the device to perform a method comprising:
speech recognizing in real-time, audio of a telephone call from a caller to a called party;
parsing text from the speech recognized audio and comparing words of the text to a hierarchical semantic tree of terms;
establishing different risk degrees for each level of a multiplicity of nodes of the hierarchical semantic tree of terms; and,
on condition that a match is detected between one of the words and a word of a node of the tree, identifying a furthest depth of the tree at which the one of the words and the term of the node match, identifying a level for the furthest depth of the tree, correlating the identified level to a degree of risk of the telephone call, and displaying an alert to the called party reflecting the degree of risk.
14. The computer program product of claim 13 , wherein the method further comprises:
submitting the speech recognized audio to a sentiment analysis engine to receive an indication of sentiment corresponding to the telephone call;
comparing the indication of sentiment to a pre-determined common sentiment for a classification of the telephone call; and,
performing the displaying only on upon the indication of sentiment differing from the common sentiment.
15. The computer program product of claim 13 , wherein the method further comprises:
submitting the speech recognized audio to a sentiment analysis engine to receive an indication of sentiment corresponding to the telephone call;
comparing the indication of sentiment to a pre-determined common sentiment for a classification of the telephone call; and,
additionally displaying a confirmation of the alert upon the indication of sentiment differing from the common sentiment.
16. The computer program product of claim 13 , wherein the method further comprises:
detecting a source phone number for the call and identifying a geographic location from which the telephone call originates;
determining if the geographic location is associated with a higher risk of a fraudulent telephone call; and,
performing the speech recognizing, parsing, identifying and displaying in response to determining that the geographic location is associated with a higher risk of a fraudulent telephone call.
17. The computer program product of claim 14 , wherein the method further comprises:
recording a set of telephone calls received by the called party;
classifying each telephone call in the set; and,
submitting each of the telephone calls in the set to a sentiment analysis engine in order to compute a sentiment common to each classification of the set.
18. The computer program product of claim 15 , wherein the method further comprises:
recording a set of telephone calls received by the called party;
classifying each telephone call in the set; and,
submitting each of the telephone calls in the set to a sentiment analysis engine in order to compute a sentiment common to each classification of the set.