Bot for a communication platform

Patent metadata

Field	Value
Publication number	US-11418462-B2
Application number	US-201816214538-A
Country	US
Kind code	B2
Filing date	Dec 10, 2018
Priority date	Dec 12, 2017
Publication date	Aug 16, 2022
Grant date	Aug 16, 2022

Abstract

Provided is a bot for a communication platform, wherein the bot is set up to read in an input data record, wherein the input data record includes a plurality of tickets or a plurality of numerical representations of tickets, listen in on at least one communication process between at least two persons on the communication platform, wherein the at least one communication process has at least one service case or at least one numerical representation of a service case, and generate an output data record, wherein the output data record has the input data record and the at least one communication process. A corresponding communication platform, a computer-implemented method, and a computer program product are also provided.

First claim

The invention claimed is:

1. A bot for a communication platform, wherein the bot is stored in a memory of the communication platform, and when executed by a processor of the communication platform causes the bot to perform the steps of:
joins an active conference between at least two persons on the communication platform regarding at least one service case, the active conference being initiated by at least one person of the at least two persons on the communication platform to identify a problem with at least one of an installation and a component associated with the at least one service case, wherein the active conference is a live chat session between at least two persons;
converts the at least one service case into a numerical representation of the at least one service case using natural language processing and machine learning;
reads in an input data record comprising a plurality of tickets of past or already present service cases, wherein the bot has converted the plurality of tickets into a plurality of numerical representations of the tickets using natural language processing and machine learning;
listens in on the live chat session between at least two persons on the communication platform, wherein information is exchanged between the at least two persons during the live chat session and is associated with the at least one service case;
wherein the bot listens in on a further communication process between the at least two persons on the communication platform, wherein the further communication process has the at least one service case;
proactively identifies the problem and offers a solution to the problem to the at least two persons during the live chat session without being prompted by the at least two persons, wherein, as a function of proactively identifying the problem and offering the solution, the bot:
accesses a database to locate a similar case that is similar to the at least one service case, a similarity value being determined by performing a similarity calculation using a similarity function that compares the numerical representation of the at least one service case with the plurality of numerical representation stored in the database, wherein the bot generates the similar case by applying a linguistic algorithm to the further communication process if the similarity value undershoots a particular threshold value;
identifies the problem associated with the at least one service and offers the solution to the problem to the at least two persons during the live chat session on the communication platform based on a problem and solution of the similar case so that a response time to the at least one service case is shortened and a downtime experienced by the at least one of an installation and the component is at least reduced;
combines the input data record with a content of the live chat session which has been listened in on during the active conference between the at least two persons; and
generates an output data record which is stored in the database, wherein the output data record has the input data record and the content of the live chat session.

2. The bot as claimed in claim 1 , wherein the similarity calculation includes determining a cosine similarity and a cosine similarity value is allocated to the at least one service case on a basis of the cosine similarity.

3. The bot as claimed in claim 1 , wherein the bot generates at least one ticket from the at least one service case.

4. A communication platform, comprising a processor and memory, wherein a bot is stored in the memory, and when executed by the processor of the communication platform causes the bot to perform the steps of:
joins an active conference between at least two persons on the communication platform regarding at least one service case, the active conference being initiated by at least one person of the at least two persons on the communication platform to identify a problem with at least one of an installation and a component associated with the at least one service case, wherein the active conference is a live chat session between at least two persons;
converts the at least one service case into a numerical representation of the at least one service case using natural language processing and machine learning;
reads in an input data record comprising a plurality of tickets of past or already present service cases, wherein the bot has converted the plurality of tickets into a plurality of numerical representations of the tickets using natural language processing and machine learning;
listens in on the live chat session between at least two persons on the communication platform, wherein information is exchanged between the at least two persons during the live chat session and is associated with the at least one service case;
wherein the bot listens in on a further communication process between the at least two persons on the communication platform, wherein the further communication process has the at least one service case;
proactively identifies the problem and offers a solution to the problem to the at least two persons during the live chat session without being prompted by the at least two persons, wherein, as a function of proactively identifying the problem and offering the solution, the bot:
accesses a database to locate a similar case that is similar to the at least one service case, a similarity value being determined by performing a similarity calculation using a similarity function that compares the numerical representation of the at least one service case with the plurality of numerical representation stored in the database, wherein the bot generates the similar case by applying a linguistic algorithm to the further communication process if the similarity value undershoots a particular threshold value;
identifies the problem associated with the at least one service and offers the solution to the problem to the at least two persons during the live chat session on the communication platform based on a problem and solution of the similar case so that a response time to the at least one service case is shortened and a downtime experienced by the at least one of an installation and the component is at least reduced;
combines the input data record with a content of the live chat session which has been listened in on during the active conference between the at least two persons; and
generates an output data record which is stored in the database, wherein the output data record has the input data record and the content of the live chat session.

5. A computer-implemented method for a communication platform comprising:
joining an active conference between at least two persons on the communication platform regarding at least one service case, the active conference being initiated by at least one person of the at least two persons on the communication platform to identify a problem with at least one of an installation and a component associated with the at least one service case, wherein the active conference is a live chat session between at least two persons;
converting the at least one service case into a numerical representation of the at least one service case using natural language processing and machine learning;
reading in an input data record by means of a bot, wherein the input data record comprises a plurality of tickets of past or already present service cases, wherein the bot has converted the plurality of tickets into a plurality of numerical representations of the tickets using natural language processing and machine learning;
listening in on the live chat session of the active conference between at least two persons on the communication platform by means of the bot, wherein information is exchanged between the at least two persons during the live chat session and is associated with the at least one serv

Assignees

• Siemens Ag

Inventors

• Hubauer Thomas
• Karlsson Christer

Classifications

• G06Q10/20 Primary

  Administration of product repair or maintenance · CPC title

• H04L67/51

  Discovery or management thereof, e.g. service location protocol [SLP] or web services · CPC title

• H04L51/02 Primary

  using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages · CPC title

• H04L51/216

  Handling conversation history, e.g. grouping of messages in sessions or threads · CPC title

• G06Q10/107

  Computer-aided management of electronic mailing [e-mailing] · CPC title