Back to results
US 11,934,977 B2Granted
Dynamic and continuous onboarding of service providers in an online expert marketplace
AssigneeTaskhuman, Inc.
InventorsRavi Swaminathan, Kartik Thumbavanam Padmanabhan
Life of this patent
Estimating the term…
Abstract
A system to generate and maintain a database of service provider skills and rankings with various categories is disclosed. Skills and rankings are generated from a number of corpus texts as well as service provider content. The database is dynamically updated to reflect changes to the corpus texts and/or service provider content.
Claims16
A method for analyzing skills of a service provider, the method comprising:
- receiving a corpus text comprising a plurality of articles;
- parsing the plurality of articles using text pre-processing, the text pre-processing comprising at least one of a stop-word removal processing, a lemmatization processing, or an entity recognition processing;
- training an online model using the parsed plurality of articles, the online model being trained using at least one of a deep learning training or a doc2vec training and the online model comprising at least one of a deep learning element, an autonomous database element, a Bayesian statistics element, a recommendation system element, or a search relevance element;
- automatically detecting a change to the plurality of articles in the corpus text has occurred;
- automatically parsing the changed plurality of articles using the text pre-processing;
- automatically training the online model using the parsed changed plurality of articles;
- automatically accessing content of the service provider;
- extracting a topic from the content using the online model;
- mapping the topic to a category of a corpus or a spot database as a vector representative of the topic;
- storing the mapped topic as the vector representative of the topic in a topics database;
- selecting the mapped topic from the topics database;
- presenting the selected topic to the service provider via a display;
- receiving user input provided by the service provider, the user input indicating approval of the selected topic;
- updating a skill rating of the service provider in a provider database for the category corresponding to the selected topic based on the approval indicated in the user input;
- determining topics that the service provider currently services based on the vector representative of the topic;
- identifying a topic that is related to the topics that the service provider currently services based on the vector representative of the topic and a vector representative of the topic that is related to the topics that the service provider currently services;
- and
- presenting the topic that is related to the topics that the service provider currently services to the service provider via the display.
15 more elements
A method to find topics of expertise to recommend to a first service provider, the method comprising:
- receiving a corpus text comprising a plurality of articles;
- parsing the plurality of articles using text pre-processing, the text pre-processing comprising at least one of a stop-word removal processing, a lemmatization processing, or an entity recognition process;
- training an online model using the parsed plurality of articles, the online model being trained using at least one of a deep learning training or a doc2vec training and the online model comprising at least one of a deep learning element, an autonomous database element, a Bayesian statistics element, a recommendation system element, or a search relevance element;
- automatically detecting a change to the plurality of articles in the corpus text has occurred;
- automatically parsing the changed plurality of articles using the text pre-processing;
- automatically training the online model using the parsed changed plurality of articles;
- accessing content of the first service provider;
- extracting one or more topics of expertise from the content using the online model;
- finding a topic in a corpus or a spot database similar to but not present in the topic of expertise of the first service provider;
- storing the found topic as a vector representative of the found topic in a topics database;
- selecting the found topic from the topics database;
- notifying the first service provider of the found topic via a display;
- receiving user input provided by the first service provider, the user input indicating approval of the found topic;
- updating a list of expertise of the first service provider in a provider database to include the found topic based on the approval indicated in the user input;
- determining topics that the service provider currently services based on the vector representative of the found topic;
- identifying a topic that is related to the topics that the service provider currently services based on the vector representative of the found topic and a vector representative of the topic that is related to the topics that the service provider currently services;
- and
- presenting the topic that is related to the topics that the service provider currently services to the service provider via the display.
15 more elements
A method to onboard a first service provider in an expert marketplace, the method comprising:
- receiving a corpus text comprising a plurality of articles;
- parsing the plurality of articles using text pre-processing, the text pre-processing comprising at least one of a stop-word removal processing, a lemmatization processing, or an entity recognition process;
- training an online model using the parsed plurality of articles, the online model being trained using at least one of a deep learning training or a doc2vec training and the online model comprising at least one of a deep learning element, an autonomous database element, a Bayesian statistics element, a recommendation system element, or a search relevance element;
- automatically detecting a change to the plurality of articles in the corpus text has occurred;
- automatically parsing the changed plurality of articles using the text pre-processing;
- automatically training the online model using the parsed changed plurality of articles;
- receiving one or more characters of text from the first service provider;
- extracting a topic from the one or more characters of text using the online model;
- generating a category by matching a vector representative of the topic to a corpus or a spot database;
- augmenting the category by searching the corpus or the spot database for a similar category;
- displaying the category and the similar category;
- wherein the corpus or the spot database contains categories created by processing content of a second service provider;
- receiving user input provided by the first service provider, the user input indicating approval at least of the category and the similar category;
- updating a list of categories associated with the first service provider in a provider database based on the approval indicated in the user input;
- determining topics that the service provider currently services based on the vector representative of the topic;
- identifying a topic that is related to the topics that the service provider currently services based on the vector representative of the topic and a vector representative of the topic that is related to the topics that the service provider currently services;
- and
- presenting the topic that is related to the topics that the service provider currently services to the service provider via the display.
15 more elements
Dependencies are inferred from the claim text.