IP Library Granted Patent US 11,388,286
Granted Patent B2
US 11,388,286 · App. 17/335,123 · Granted Jul 12, 2022

System and method for handling unwanted telephone calls

Inventors: Dmitry V. Shvetsov (Moscow, RU); Daniil A. Yazovsky (Moscow, RU); Vitaly S. Vorobiov (Moscow, RU)
Assignee: AO Kaspersky Lab
H04M3/436G06N20/10G10L15/14H04M3/2218H04M3/5307
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,388,286
App. No.
17/335,123
Granted
Jul 12, 2022
Kind
B2
Abstract

Disclosed herein are systems and methods for handling unwanted telephone calls. In one aspect, an exemplary method comprises, intercepting a call request for a call from a terminal device of a calling party to a terminal device of a called party, generating a call recording containing media data transmitted within a connection established by the intercepted call request, determining attributes of the generated call recording, classifying the call as an unwanted call based on the determined attributes, wherein the classification is performed by a classifier trained on previously collected unwanted calls, and wherein the call is classified as unwanted when the attributes belong to an unwanted call class that is known, and handling the call in accordance with the classification of the call, the handling including at least securing information of the call.

Claims (64)

1. A method for handling unwanted telephone calls, the method comprising:

intercepting a call request for a call from a terminal device of a calling party to a terminal device of a called party;

generating a call recording containing media data transmitted within a connection established by the intercepted call request, wherein the call recording is generated over a specified time interval at which a size of the media data is a multiple of a preset value;

determining attributes of the generated call recording;

classifying the call as an unwanted call based on the determined attributes, wherein the classification is performed by a classifier trained on previously collected unwanted calls, and wherein the call is classified as unwanted when the attributes belong to an unwanted call class that is known; and

handling the call in accordance with the classification of the call, the handling including at least securing information of the call.

2. The method of claim 1 , wherein an attribute of the generated call recording consists of at least one of:

a set of words;

a vector representation of words; and

a multiset of words disregarding grammar and word order.

3. The method of claim 1 , wherein the attributes of the generated call recording are determined using an automatic speech recognizer.

4. The method of claim 1 , wherein classes of the call include at least:

the unwanted call class; and

a genuine call class.

5. The method of claim 1 , wherein the call is classified as unwanted when the call is initiated by the calling party to carry out at least one of:

a fraud;

an obtaining of confidential information; and

offering of goods that cause harm to health and/or development of the called party.

6. The method of claim 1 , wherein the classification is performed by the classifier trained using a call classification model and the previously collected unwanted calls.

7. The method of claim 1 , wherein an algorithm of the call classifier comprises at least an algorithm based on:

Bayesian classifiers;

a logistical regression;

a Markov Random Field (MRF) classifier;

a support vector machine;

methods using nearest neighbors; and

a decision tree.

8. The method of claim 1 , wherein the call classification model is trained on attributes of any number of previously classified calls, and wherein the class of the previously classified calls have been confirmed in such a way that a subsequent call is less likely to be incorrectly classified than a current call.

9. The method of claim 1 , wherein the call is handled at least by one of the following:

an early termination of the call; and

informing the called party that the call belongs to the unwanted call class.

10. A system for handling unwanted telephone calls, comprising:

at least one processor configured to:

intercept a call request for a call from a terminal device of a calling party to a terminal device of a called party;

generate a call recording containing media data transmitted within a connection established by the intercepted call request, wherein the call recording is generated over a specified time interval at which a size of the media data is a multiple of a preset value;

determine attributes of the generated call recording;

classify the call as an unwanted call based on the determined attributes, wherein the classification is performed by a classifier trained on previously collected unwanted calls, and wherein the call is classified as unwanted when the attributes belong to an unwanted call class that is known; and

handle the call in accordance with the classification of the call, the handling including at least securing information of the call.

11. The system of claim 10 , wherein an attribute of the generated call recording consists of at least one of:

a set of words;

a vector representation of words; and

a multiset of words disregarding grammar and word order.

12. The system of claim 10 , wherein the attributes of the generated call recording are determined using an automatic speech recognizer.

13. The system of claim 10 , wherein classes of the call include at least:

the unwanted call class; and

a genuine call class.

14. The method of claim 10 , wherein the call is classified as unwanted when the call is initiated by the calling party to carry out at least one of:

a fraud;

an obtaining of confidential information; and

offering of goods that cause harm to health and/or development of the called party.

15. The system of claim 10 , wherein the classification is performed by the classifier trained using a call classification model and the previously collected unwanted calls.

16. The system of claim 10 , wherein an algorithm of the call classifier comprises at least an algorithm based on:

Bayesian classifiers;

a logistical regression;

a Markov Random Field (MRF) classifier;

a support vector machine;

methods using nearest neighbors; and

a decision tree.

17. The system of claim 10 , wherein the call classification model is trained on attributes of any number of previously classified calls, and wherein the class of the previously classified calls have been confirmed in such a way that a subsequent call is less likely to be incorrectly classified than a current call.

18. A non-transitory computer readable medium storing thereon computer executable instructions for handling unwanted telephone calls, including instructions for:

intercepting a call request for a call from a terminal device of a calling party to a terminal device of a called party;

generating a call recording containing media data transmitted within a connection established by the intercepted call request, wherein the call recording is generated over a specified time interval at which a size of the media data is a multiple of a preset value;

determining attributes of the generated call recording;

classifying the call as an unwanted call based on the determined attributes, wherein the classification is performed by a classifier trained on previously collected unwanted calls, and wherein the call is classified as unwanted when the attributes belong to an unwanted call class that is known; and

handling the call in accordance with the classification of the call, the handling including at least securing information of the call.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 1, 2021
From: SHVETSOV, DMITRY V; YAZOVSKY, DANIIL A; VOROBIOV, VITALY S
To: AO KASPERSKY LAB
Reel/Frame 056397/0549 →
Priority Claims (1)
RU RU2020131448 · Sep 24, 2020 · national
Continuity (1)
Related Publication 20220094784A1 · Mar 24, 2022