AI communication management system and method based on mobile communication equipment

Abstract

The present invention discloses an AI communication management system and method based on mobile communication equipment, which relates to the technical field of communication record management, performs text recognition on the voice of the communication record, extracts characteristic words from the text recognition result, vectorizes the characteristic words, identifies the keywords in each communication record through clustering analysis on the word vectors, gathers into a set of related words corresponding to the keywords according to the clustering result, obtains the corresponding call record of a certain set of related words, performs feature extraction on the call record, obtains the subject words of the call record, obtains the keywords corresponding to a certain set of related words, obtains the real-time call record of the user, identifies the keywords in the real-time call record, obtains all voice communication records including the keywords, obtains the associated word set of each voice communication record respectively, and feeds back the voice communication record corresponding to the associated word set to the user.

Description

Technical Field

[0001] The present invention relates to the technical field of communication record management, and in particular to an AI communication management system and method based on mobile communication equipment. Background Art

[0002] With the development of information technology, the way of delivering documents has gradually changed from face-to-face delivery to delivery through an online platform. The relevant staff uploads the electronic version of the document to the online platform, and the person to be served downloads the corresponding document from the online platform. Due to the lack of face-to-face communication between the relevant staff and the person to be served, there are cases where the relevant staff and the person to be served communicate on the content of the document by telephone. The relevant staff needs to contact the same person to be served multiple times. During each contact, the focus of the communication content between the relevant staff and the person to be served is different. At the same time, the time span of the telephone communication is relatively large. Therefore, the relevant staff needs to record the communication with the person to be served and search for the record during the communication process with the person to be served. There are omissions or forgetfulness in the process of recording the content of the call and searching for the record, which brings inconvenience to the work of the relevant staff. Summary of the invention

[0003] The purpose of the present invention is to provide an AI communication management system and method based on mobile communication devices to solve the problems raised in the prior art.

[0004] To achieve the above object, the present invention provides the following technical solution: an AI communication management method based on a mobile communication device, the method comprising:

[0005] Step S100: the user logs in user information on two communication devices, obtains the document and corresponding contact information from the document delivery platform through one of the two communication devices, conducts voice communication with the contact through the other communication device, and records the process of the voice communication. After the voice communication is completed, the communication record is saved;

[0006] Step S200: Perform text recognition on the speech in the communication record, extract feature words from the text recognition results, vectorize the feature words, identify the keywords in each communication record through clustering analysis of the word vectors, and store these keywords in the database. According to the clustering results, obtain the related words of each keyword and collect them into the related word set corresponding to the keyword;

[0007] Step S300: Obtain call records corresponding to a certain set of related words, perform feature extraction on the call records, obtain subject words of the call records, obtain keywords corresponding to a certain set of related words, and group the subject words into a set of associated words of the keywords;

[0008] Step S400: obtaining the user's real-time call records, identifying keywords in the real-time call records, obtaining all voice communication records including the keywords, obtaining the associated word sets of each voice communication record respectively, obtaining the associated word set selected by the user, and feeding back the voice communication records corresponding to the associated word set to the user.

[0009] Furthermore, step S100 includes:

[0010] Step S101: Obtain the correspondence between the document and the contact person, as well as the contact information of the contact person, from the document delivery platform, and establish a management database for the contact information of the document and the contact person;

[0011] Step S102: The user logs in user information on two devices at the same time, the two devices are respectively recorded as a first device and a second device, wherein the first device includes a network communication device, the first device includes an information display interface and a human-computer interaction interface, and the second device includes a voice communication device;

[0012] Step S103: taking a document selected by the user as a target document, obtaining a contact person of the target document, and recording the contact person as a target contact person;

[0013] Step S104: the user obtains the contact information of the target contact on the document through the first device, pushes the contact information to the user's second device, and the user makes a voice call with the target contact through the second device;

[0014] Step S105: The second device records the time information of the call record, the time information including the start time and the end time of the call record, performs voice recognition on the call record between the user and the target contact, and saves the call record as a voice record and a text record;

[0015] Step S106: Feedback the voice record, text record and time information to the first device.

[0016] Furthermore, the method for the second device to obtain the contact information of the target contact includes:

[0017] a: Displaying the contact information of the target contact in the form of text or picture through the first device, and when the user sees the text or picture information, the user contacts the target contact;

[0018] b: The contact information of the target contact is stored in the network address associated with the QR code. The user clicks the relevant link in the first device to display the QR code. The user identifies the QR code through the second device to obtain the contact information of the target contact.

[0019] c: The user operates the human-computer interaction interface in the first device to send the contact information of the target contact to the second device in the form of data transmission.

[0020] Further, step S200 includes:

[0021] Step S201: extracting feature words from the text record, recording the feature words as first feature words, and collecting the first feature words in the text record corresponding to a voice call into a first feature word set;

[0022] Step S202: performing word vectorization on a first feature word in a feature word set to obtain a word vector corresponding to the first feature word;

[0023] Step S203: In a text record corresponding to a voice call, cluster the word vectors in a first feature word set, use the first feature word at the cluster center as a keyword, and aggregate the keywords into a keyword database;

[0024] Step S204: according to the classification relationship of the first characteristic words obtained by clustering, the first characteristic words in each cluster are respectively collected and recorded in a related word set of the corresponding keyword;

[0025] The words are vectorized and the communication records are segmented. The keywords are extracted and recorded as the central words of each segment. By extracting the central words and sorting them by time, that is, arranging the segments of the communication records, the integration of communication records with similar semantics in different time periods is improved, thereby improving the efficiency of information retrieval.

[0026] Furthermore, step S300 includes:

[0027] Step S301: Obtain time information of the call record and a certain related word set, obtain the time when each feature word in the related word set appears in the call record, and arrange the feature words according to the order of the time;

[0028] Step S302: obtaining the minimum time value tmin and the maximum time value tmax, obtaining the time range from tmin to tmax, and taking all text records within the time range as target records;

[0029] Step S303: Acquire feature words in the target record, record the feature words as second feature words, collect the second feature words, and obtain a second feature word set of keywords;

[0030] Step S304: Count the frequency of each second characteristic word in the target record, arrange the second characteristic words in the second characteristic word set in descending order of frequency, select the first k second characteristic words in descending order of frequency, and compile them into the associated word set recorded as the keyword;

[0031] In order to distinguish the differences between different fragments of the same keyword, the fragments corresponding to the same keyword extract the second feature word;

[0032] The second feature word represents the main content involved in the call records within a continuous time range. In order to highlight the characteristics of the same keyword in different contexts, the second feature word is used to represent the relevant information of the keyword in different contexts, which facilitates users to select corresponding historical records.

[0033] Furthermore, step S400 includes:

[0034] Step S401: Recording a real-time call between a user and a target contact through a second device to obtain real-time record information of the call;

[0035] Step S402: searching for keywords in the real-time recorded information, recording the searched keywords as target keywords, and pushing the target keywords to the first device;

[0036] Step S403: Acquire all keyword sets corresponding to the target keyword, arrange the keyword sets according to the distance from the real-time call time from near to far, and display the arranged keyword sets to the user on the first device;

[0037] Step S404: taking the keyword set selected by the user in the first device as the target keyword set, recording the time range corresponding to the target keyword set as the target time range, obtaining all text records in the target time range, and displaying them to the user on the first device.

[0038] In order to better implement the above method, an AI communication management system based on mobile communication equipment is also proposed, the system includes:

[0039] A call record management module, a keyword management module, an associated word management module and a real-time monitoring module, wherein the call record management module is used to manage the user's first device and the second device, the keyword management module is used to manage the keywords and the related words of the associated words, the associated word management module is used to obtain the position of the keyword in the text record, extract the associated words at the position to form an associated word set, and the real-time monitoring module is used to obtain the user's real-time call record, obtain the relevant information of the historical call record and the real-time call record, and push it to the user;

[0040] Further, the call record management module includes: a document management unit, a voice recognition unit and a record management unit, wherein the document management unit is used to manage a management database of documents and contact information of contacts, the voice recognition unit is used to recognize text information in call records, and the record management unit is used to manage the correspondence between voice records, text records and time information of call records;

[0041] Further, the keyword management module includes: a first feature extraction unit, a cluster analysis unit, a keyword management unit and a related word management unit, wherein the first feature extraction unit is used to obtain a first feature word in a text record, the cluster analysis unit is used to vectorize the first feature word, and perform cluster analysis on the word vector, the keyword management unit is used to manage a keyword database, and the related word management unit is used to manage a set of related words corresponding to the keyword;

[0042] Further, the associated word management module includes: a target record management unit, a second feature extraction unit, an associated word management unit and a word frequency management unit, wherein the target record management unit is used to obtain a target record corresponding to the keyword, the second feature extraction unit is used to obtain a second feature word of the target record, the associated word management unit is used to manage an associated word set of the keyword, and the word frequency management unit is used to perform word frequency statistics on the second feature word in the associated word set;

[0043] Furthermore, the real-time monitoring module includes: a call record unit, a keyword recognition unit, a related word display unit and a text record acquisition unit, wherein the call record unit is used to manage real-time record information of calls, the keyword recognition unit is used to identify keywords in real-time call records, the related word display unit is used to display keywords and corresponding associated word sets to users, and the text record acquisition unit is used to obtain historical records related to the associated word set selected by the user.

[0044] Compared with the prior art, the invention has the following beneficial effects: extracting keyword information from call records, and summarizing and marking differences of historical records related to the same keyword by classifying and sorting the call records. On the one hand, it overcomes the omission of key points in the call records during manual recording, and on the other hand, it is convenient to provide users with an intuitive selection scheme, so that users can select the required historical records while communicating on the phone, thereby improving the work efficiency of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a schematic diagram of the structure of the AI ​​communication management system based on mobile communication devices of the present invention;

[0046] Figure 2 The figure is a flow chart of the AI ​​communication management method based on mobile communication equipment of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] Example: Figure 1 and Figure 2 As shown, the present invention provides a technical solution, an AI communication management system and method based on a mobile communication device, the method comprising:

[0049] Step S100: the user logs in user information on two communication devices, obtains the document and corresponding contact information from the document delivery platform through one of the two communication devices, conducts voice communication with the contact through the other communication device, and records the process of the voice communication. After the voice communication is completed, the communication record is saved;

[0050] Wherein, step S100 includes:

[0051] Step S101: Obtain the correspondence between the document and the contact person, as well as the contact information of the contact person, from the document delivery platform, and establish a management database for the contact information of the document and the contact person;

[0052] Step S102: The user logs in user information on two devices at the same time, the two devices are respectively recorded as a first device and a second device, wherein the first device includes a network communication device, the first device includes an information display interface and a human-computer interaction interface, and the second device includes a voice communication device;

[0053] Step S103: taking a document selected by the user as a target document, obtaining a contact person of the target document, and recording the contact person as a target contact person;

[0054] Step S104: the user obtains the contact information of the target contact on the document through the first device, pushes the contact information to the user's second device, and the user makes a voice call with the target contact through the second device;

[0055] Step S105: The second device records the time information of the call record, the time information including the start time and the end time of the call record, performs voice recognition on the call record between the user and the target contact, and saves the call record as a voice record and a text record;

[0056] Step S106: Feedback the voice record, text record and time information to the first device;

[0057] In an embodiment, the first device, such as a computer device, implements information sending and historical record search by operating a visualized computer interface, and the second device, such as a smart phone device, obtains the contact information of the target contact through information reception and intelligent recognition technology;

[0058] In an embodiment, the document delivery platform is associated with an independent address book, or uses an independent communication device to distinguish it from a private address book or private communication device;

[0059] With the user's authorization, relevant address books and call records of calls with relevant address book contacts are collected.

[0060] The method for the second device to obtain the contact information of the target contact includes:

[0061] a: Displaying the contact information of the target contact in the form of text or picture through the first device, and when the user sees the text or picture information, the user contacts the target contact;

[0062] b: The contact information of the target contact is stored in the network address associated with the QR code. The user clicks the relevant link in the first device to display the QR code. The user identifies the QR code through the second device to obtain the contact information of the target contact.

[0063] c: The user operates the human-computer interaction interface in the first device to send the contact information of the target contact to the second device in the form of data transmission.

[0064] Step S200: Perform text recognition on the speech in the communication record, extract feature words from the text recognition results, vectorize the feature words, identify the keywords in each communication record through clustering analysis of the word vectors, and store these keywords in the database. According to the clustering results, obtain the related words of each keyword and collect them into the related word set corresponding to the keyword;

[0065] Wherein, step S200 includes:

[0066] Step S201: extracting feature words from the text record, recording the feature words as first feature words, and collecting the first feature words in the text record corresponding to a voice call into a first feature word set;

[0067] Step S202: performing word vectorization on a first feature word in a feature word set to obtain a word vector corresponding to the first feature word;

[0068] Step S203: In a text record corresponding to a voice call, cluster the word vectors in a first feature word set, use the first feature word at the cluster center as a keyword, and aggregate the keywords into a keyword database;

[0069] Step S204: According to the classification relationship of the first characteristic words obtained by clustering, the first characteristic words in each cluster are respectively collected and recorded in the related word set of the corresponding keyword.

[0070] Step S300: Obtain call records corresponding to a certain set of related words, perform feature extraction on the call records, obtain subject words of the call records, obtain keywords corresponding to a certain set of related words, and group the subject words into a set of associated words of the keywords;

[0071] Wherein, step S300 includes:

[0072] Step S301: Obtain time information of the call record and a certain related word set, obtain the time when each feature word in the related word set appears in the call record, and arrange the feature words according to the order of the time;

[0073] Step S302: obtaining the minimum time value tmin and the maximum time value tmax, obtaining the time range from tmin to tmax, and taking all text records within the time range as target records;

[0074] Step S303: Acquire feature words in the target record, record the feature words as second feature words, collect the second feature words, and obtain a second feature word set of keywords;

[0075] Step S304: Count the frequency of each second characteristic word in the target record, arrange the second characteristic words in the second characteristic word set in descending order of frequency, select the first k second characteristic words in descending order of frequency, and compile them into the associated word set recorded as the keyword;

[0076] By generating different associated words for the same keyword for the user to choose from, the system can correctly associate the historical call records required by the user.

[0077] Step S400: obtaining the user's real-time call records, identifying keywords in the real-time call records, obtaining all voice communication records including the keywords, respectively obtaining the associated word sets of each voice communication record, obtaining the associated word set selected by the user, and feeding back the voice communication records corresponding to the associated word set to the user;

[0078] Wherein, step S400 includes:

[0079] Step S401: Recording a real-time call between a user and a target contact through a second device to obtain real-time record information of the call;

[0080] Step S402: searching for keywords in the real-time recorded information, recording the searched keywords as target keywords, and pushing the target keywords to the first device;

[0081] Step S403: Acquire all keyword sets corresponding to the target keyword, arrange the keyword sets according to the distance from the real-time call time from near to far, and display the arranged keyword sets to the user on the first device;

[0082] Step S404: taking the keyword set selected by the user in the first device as the target keyword set, recording the time range corresponding to the target keyword set as the target time range, obtaining all text records in the target time range, and displaying them to the user on the first device.

[0083] The system includes:

[0084] Call record management module, keyword management module, associated word management module and real-time monitoring module;

[0085] The call record management module is used to manage the first device and the second device of the user, wherein the call record management module includes: a document management unit, a voice recognition unit and a record management unit, wherein the document management unit is used to manage a management database of documents and contact information of contacts, the voice recognition unit is used to recognize text information in the call record, and the record management unit is used to manage the correspondence between the voice record, text record and time information of the call record;

[0086] The keyword management module is used to manage the related words of the keywords and associated words, wherein the keyword management module includes: a first feature extraction unit, a cluster analysis unit, a keyword management unit and a related word management unit, wherein the first feature extraction unit is used to obtain the first feature word in the text record, the cluster analysis unit is used to vectorize the first feature word, and perform cluster analysis on the word vector, the keyword management unit is used to manage the keyword database, and the related word management unit is used to manage the related word set corresponding to the keyword;

[0087] The associated word management module is used to obtain the position of the keyword in the text record, extract the associated words at the position to form an associated word set, wherein the associated word management module includes: a target record management unit, a second feature extraction unit, an associated word management unit and a word frequency management unit, wherein the target record management unit is used to obtain the target record corresponding to the keyword, the second feature extraction unit is used to obtain the second feature word of the target record, the associated word management unit is used to manage the associated word set of the keyword, and the word frequency management unit is used to perform word frequency statistics on the second feature word in the associated word set;

[0088] Among them, the real-time monitoring module is used to obtain the user's real-time call records, obtain relevant information of historical call records and real-time call records, and push them to the user. Among them, the real-time monitoring module includes: a call record unit, a keyword recognition unit, a related word display unit and a text record acquisition unit. Among them, the call record unit is used to manage the real-time record information of the call, the keyword recognition unit is used to identify the keywords in the real-time call records, the related word display unit is used to display the keywords and the corresponding associated word sets to the user, and the text record acquisition unit is used to obtain historical records related to the associated word set selected by the user.

[0089] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and range of equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. An AI communication management method based on mobile communication equipment, characterized in that: The method comprises the following steps: Step S100: The user logs in user information on two communication devices, obtains the document and corresponding contact information from the document delivery platform through one of the two communication devices, conducts voice communication with the contact through the other communication device, and records the process of the voice communication. After the voice communication is completed, the communication record is saved; Step S200: Perform text recognition on the speech in the communication record, extract feature words from the text recognition results, vectorize the feature words, identify the keywords in each communication record through clustering analysis of the word vectors, and store these keywords in the database. According to the clustering results, obtain the related words of each keyword and collect them into the related word set corresponding to the keyword; Step S300: obtaining call records corresponding to a certain set of related words, performing feature extraction on the call records, obtaining subject words of the call records, obtaining keywords corresponding to the certain set of related words, and grouping the subject words into a set of associated words of the keywords; Step S400: obtaining the user's real-time call records, identifying keywords in the real-time call records, obtaining all voice communication records including the keywords, respectively obtaining the associated word sets of each voice communication record, obtaining the associated word set selected by the user, and feeding back the voice communication records corresponding to the associated word set to the user.

2. The AI ​​communication management method based on mobile communication equipment according to claim 1, characterized in that: Step S100 includes: Step S101: Obtain the correspondence between the document and the contact person, as well as the contact information of the contact person, from the document delivery platform, and establish a management database for the contact information of the document and the contact person; Step S102: The user logs in user information on two devices at the same time, the two devices are respectively recorded as a first device and a second device, wherein the first device includes a network communication device, the first device includes an information display interface and a human-computer interaction interface, and the second device includes a voice communication device; Step S103: taking a document selected by the user as a target document, obtaining a contact person of the target document, and recording the contact person as a target contact person; Step S104: the user obtains the contact information of the target contact on the document through the first device, pushes the contact information to the second device of the user, and the user makes a voice call with the target contact through the second device; Step S105: The second device records the time information of the call record, the time information including the start time and the end time of the call record, performs voice recognition on the call record between the user and the target contact, and saves the call record as a voice record and a text record; Step S106: Feedback the voice record, text record and time information to the first device.

3. The AI ​​communication management method based on mobile communication equipment according to claim 2, characterized in that: The method for the second device to obtain the contact information of the target contact includes: a: Displaying the contact information of the target contact in the form of text or picture through the first device, and when the user sees the text or picture information, the user contacts the target contact; b: The contact information of the target contact is stored in the network address associated with the QR code. The user clicks the relevant link in the first device to display the QR code. The user identifies the QR code through the second device to obtain the contact information of the target contact. c: The user operates the human-computer interaction interface in the first device to send the contact information of the target contact to the second device in the form of data transmission.

4. The AI ​​communication management method based on mobile communication equipment according to claim 2, characterized in that: Step S200 includes: Step S201: extracting feature words from a text record, recording the feature words as first feature words, and collecting first feature words in a text record corresponding to a voice call into a first feature word set; Step S202: performing word vectorization on a first feature word in a feature word set to obtain a word vector corresponding to the first feature word; Step S203: In a text record corresponding to a voice call, cluster the word vectors in a first feature word set, use the first feature word at the cluster center as a keyword, and aggregate the keywords into a keyword database; Step S204: According to the classification relationship of the first characteristic words obtained by clustering, the first characteristic words in each cluster are respectively collected and recorded in the related word set of the corresponding keyword.

5. The AI ​​communication management method based on mobile communication equipment according to claim 4, characterized in that: Step S300 includes: Step S301: Obtain time information of the call record and a certain related word set, obtain the time when each feature word in the related word set appears in the call record, and arrange the feature words according to the order of the time; Step S302: obtaining a minimum time value tmin and a maximum time value tmax, obtaining a time range from tmin to tmax, and taking all text records within the time range as target records; Step S303: Acquire feature words in the target record, record the feature words as second feature words, and collect the second feature words to obtain a second feature word set of keywords; Step S304: Count the frequency of each second feature word in the target record, arrange the second feature words in the second feature word set in descending order of frequency, select the first k second feature words in descending order of frequency, and compile them into the associated word set of the keyword.

6. The AI ​​communication management method based on mobile communication equipment according to claim 5, characterized in that: Step S400 includes: Step S401: Recording a real-time call between a user and a target contact through a second device to obtain real-time record information of the call; Step S402: searching for keywords in the real-time record information, recording the searched keywords as target keywords, and pushing the target keywords to the first device; Step S403: Acquire all keyword sets corresponding to the target keyword, arrange the keyword sets according to the distance from the real-time call time from near to far, and display the arranged keyword sets to the user on the first device; Step S404: taking the keyword set selected by the user in the first device as the target keyword set, recording the time range corresponding to the target keyword set as the target time range, obtaining all text records in the target time range, and displaying them to the user on the first device.

7. An AI communication management system based on a mobile communication device, used to execute the AI ​​communication management method based on a mobile communication device according to any one of claims 1 to 6, characterized in that: The system includes: A call record management module, a keyword management module, an associated word management module and a real-time monitoring module, wherein the call record management module is used to manage the user's first device and the second device, the keyword management module is used to manage related words of keywords and associated words, the associated word management module is used to obtain the position of the keyword in the text record, extract the associated words at the position to form an associated word set, and the real-time monitoring module is used to obtain the user's real-time call record, obtain relevant information between the historical call record and the real-time call record, and push it to the user.

8. The AI ​​communication management system based on mobile communication devices according to claim 7, characterized in that: The call record management module includes: a document management unit, a voice recognition unit and a record management unit, wherein the document management unit is used to manage the management database of documents and contact information of contacts, the voice recognition unit is used to identify text information in call records, and the record management unit is used to manage the correspondence between voice records, text records and time information of call records.

9. The AI ​​communication management system based on mobile communication devices according to claim 7, characterized in that: The keyword management module includes: a first feature extraction unit, a cluster analysis unit, a keyword management unit and a related word management unit, wherein the first feature extraction unit is used to obtain a first feature word in a text record, the cluster analysis unit is used to vectorize the first feature word, and perform cluster analysis on the word vector, the keyword management unit is used to manage a keyword database, and the related word management unit is used to manage a set of related words corresponding to the keyword; The associated word management module includes: a target record management unit, a second feature extraction unit, an associated word management unit and a word frequency management unit, wherein the target record management unit is used to obtain the target record corresponding to the keyword, the second feature extraction unit is used to obtain the second feature word of the target record, the associated word management unit is used to manage the associated word set of the keyword, and the word frequency management unit is used to perform word frequency statistics on the second feature word in the associated word set.

10. The AI ​​communication management system based on mobile communication devices according to claim 7, characterized in that: The real-time monitoring module includes: a call record unit, a keyword recognition unit, a related word display unit and a text record acquisition unit, wherein the call record unit is used to manage the real-time record information of the call, the keyword recognition unit is used to identify the keywords in the real-time call record, the related word display unit is used to display the keywords and the corresponding associated word sets to the user, and the text record acquisition unit is used to obtain the historical records related to the associated word set selected by the user.

Images