Smartphone indicator for conversation nonproductivity

• US 9,722,965 B2
• Filed: 01/29/2015
• Issued: 08/01/2017
• Est. Priority Date: 01/29/2015
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Pickover, Clifford A., Cecchi, Guillermo A., Kozloski, James R., Rish, Irina
• Primary Examiner(s): Chea, Philip
• Assistant Examiner(s): Ma, Wing
• Application Number:
  US14/608,393
• Publication Number:
  US 20160226813A1
• Time in Patent Office: 915 Days
• Field of Search: 709206
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/36 Creation of semantic tools, e.g. ontology or thesauri
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: H04L 12/1895 for short real-time information, e.g. alarms, notifications, alerts, updates
• CPC Class Codes: H04L 51/04 Real-time or near real-time messaging, e.g. instant messaging [IM]
• CPC Class Codes: H04L 51/234 for tracking messages
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

A method to send an alert for nonproductivity associated with a conversation is provided. The method may include recording a plurality of communication outputs of at least two users engaged in a remote message exchange or a remote conversation. The method may also include creating a plurality of text tokens based on the recorded plurality of communication outputs. The method may include analyzing, by a graphical text analyzer, the created plurality of text tokens to determine whether the plurality of text tokens has fallen below a threshold. The method may further include sending an alert to the plurality of users involved in the conversation if it is determined that the plurality of text tokens has fallen below the threshold.

Claims

1. A method to send an alert for nonproductivity associated with an online conversation, the method comprising:

recording a plurality of communication outputs of at least two users engaged in a remote message exchange or a remote conversation, wherein the remote message exchange and the remote conversation are associated with an online communication system on a first mobile device and a second mobile device;

creating a plurality of text tokens based on the recorded plurality of communication outputs;

analyzing, by a graphical text analyzer, the created plurality of text tokens to determine the plurality of text tokens has fallen below a threshold, wherein determining if the plurality of text tokens has fallen below the threshold comprises comparing a plurality of graphs associated with the remote message exchange and the remote conversation to a prior plurality of graphs associated with a plurality of prior remote message exchanges and prior conversations;

wherein the analyzing further comprises utilizing a speaker identity feature extractor, a syntactic feature extractor, and a semantic feature extractor to analyze the recorded plurality of communication outputs;

using a graph constructor and a graph feature extractor, in conjunction with results from the speaker identity feature extractor, the syntactic feature extractor, and the semantic feature extractor to create a hybrid graph, wherein the hybrid graph comprises nodes representing words or phrases, edges representing temporal precedence in speech, and each node includes a feature vector comprising syntactic and semantic vectors, plus additional non-textual features;

constructing, by a learning engine, a predictive model or classifier using the hybrid graph; and

sending an alert to the plurality of users involved in the conversation when it is determined that the plurality of text tokens has fallen below the threshold, wherein the sent alert comprises of a color, a flashing light, and a sound that is associated with the first mobile device or the second mobile device.

2. The method of claim 1, wherein the plurality of text tokens are created to categorize the remote message exchange or the remote conversation and a cognitive state associated with at least one of the at least two users.

3. The method of claim 1, wherein the analyzing comprises comparing and clustering at least one category associated with at least one of the at least two users to create a plurality of subsequent categories.

4. The method of claim 1, wherein the analyzing comprises of a plurality of features vector that includes an identity of at least one of the at least two users and words or tokens spoken by at least one of the at least two users.

5. The method of claim 1, wherein the analyzing of the created plurality of text tokens comprises the use of a plurality of machine learning tools.

6. The method of claim 5, wherein the plurality machine learning tools are used to extract a plurality of predictive features from a plurality of token sequences to make inferences about a category of a current cognitive category of at least one of the two users and a cognitive state of the remote message exchange or the remote conversation.

7. A computer system to send an alert for nonproductivity associated with an online conversation, the computer system comprising:

one or more processors, one or more computer-readable memories, one or more computer-readable tangible storage devices, and program instructions stored on at least one of the one or more storage devices for execution by at least one of the one or more processors via at least one of the one or more memories, wherein the computer system is capable of performing a method comprising;

recording a plurality of communication outputs of at least two users engaged in a remote message exchange or a remote conversation, wherein the remote message exchange and the remote conversation are associated with an online communication system on a first mobile device and a second mobile device;

creating a plurality of text tokens based on the recorded plurality of communication outputs;

analyzing, by a graphical text analyzer, the created plurality of text tokens to determine the plurality of text tokens has fallen below a threshold, wherein determining if the plurality of text tokens has fallen below the threshold comprises comparing a plurality of graphs associated with the remote message exchange and the remote conversation to a prior plurality of graphs associated with a plurality of prior remote message exchanges and prior conversations;

wherein the analyzing further comprises utilizing a speaker identity feature extractor, a syntactic feature extractor, and a semantic feature extractor to analyze the recorded plurality of communication outputs;

using a graph constructor and a graph feature extractor, in conjunction with results from the speaker identity feature extractor, the syntactic feature extractor, and the semantic feature extractor to create a hybrid graph, wherein the hybrid graph comprises nodes representing words or phrases, edges representing temporal precedence in speech, and each node includes a feature vector comprising syntactic and semantic vectors, plus additional non-textual features;

constructing, by a learning engine, a predictive model or classifier using the hybrid graph; and

sending an alert to the plurality of users involved in the conversation when it is determined that the plurality of text tokens has fallen below the threshold, wherein the sent alert comprises of a color, a flashing light, and a sound that is associated with the first mobile device or the second mobile device.

8. The computer system of claim 7, wherein the plurality of text tokens are created to categorize the remote message exchange or the remote conversation and a cognitive state associated with at least one of the at least two users.

9. The computer system of claim 7, wherein the analyzing comprises comparing and clustering at least one category associated with at least one of the at least two users to create a plurality of subsequent categories.

10. The computer system of claim 7, wherein the analyzing comprises of a plurality of features vector that includes an identity of at least one of the at least two users and words or tokens spoken by at least one of the at least two users.

11. The computer system of claim 7, wherein the analyzing of the created plurality of text tokens comprises the use of a plurality of machine-learning tools.

12. The computer system of claim 11, wherein the plurality machine learning tools are used to extract a plurality of predictive features from a plurality of token sequences to make inferences about a category of a current cognitive category of at least one of the two users and a cognitive state of the remote message exchange or the remote conversation.

13. A computer program product to send an alert for nonproductivity associated with an online conversation, the computer program product comprising:

one or more computer-readable storage devices and program instructions stored on at least one of the one or more tangible storage devices, the program instructions executable by a processor, the program instructions comprising;

program instructions to record a plurality of communication outputs of at least two users engaged in a remote message exchange or a remote conversation, wherein the remote message exchange and the remote conversation are associated with an online communication system on a first mobile device and a second mobile device;

program instructions to create a plurality of text tokens based on the recorded plurality of communication outputs;

program instructions to analyze, by a graphical text analyzer, the created plurality of text tokens to determine the plurality of text tokens has fallen below a threshold, wherein determining if the plurality of text tokens has fallen below the threshold comprises comparing a plurality of graphs associated with the remote message exchange and the remote conversation to a prior plurality of graphs associated with a plurality of prior remote message exchanges and prior conversations;

wherein the analyzing further comprises program instructions to utilize a speaker identity feature extractor, a syntactic feature extractor, and a semantic feature extractor to analyze the recorded plurality of communication outputs;

program instructions to use a graph constructor and a graph feature extractor, in conjunction with results from the speaker identity feature extractor, the syntactic feature extractor, and the semantic feature extractor to create a hybrid graph, wherein the hybrid graph comprises nodes representing words or phrases, edges representing temporal precedence in speech, and each node includes a feature vector comprising syntactic and semantic vectors, plus additional non-textual features;

program instructions to construct, by a learning engine, a predictive model or classifier using the hybrid graph; and

program instructions to send an alert to the plurality of users involved in the conversation when it is determined that the plurality of text tokens has fallen below the threshold, wherein the sent alert comprises of a color, a flashing light, and a sound that is associated with the first mobile device or the second mobile device.

14. The computer program product of claim 13, wherein the plurality of text tokens are created to categorize the remote message exchange or the remote conversation and a cognitive state associated with at least one of the at least two users.

15. The computer program product of claim 13, wherein the analyzing comprises comparing and clustering at least one category associated with at least one of the at least two users to create a plurality of subsequent categories.