High risk passage automation in a digital transaction management platform
A document execution engine receives a training set of data including training documents that each include one or more passages associated with a passage type and a level of risk. The document execution engine trains a machine learned model based on the training set. The trained machine learned model, when applied to subsequently identified passages within documents in the document execution environment, can identify a passage with above threshold levels of risk (e.g., a high-risk passage) based on a passage type of the passage. The trained machine learned model can then provide for display the high-risk passage and a related passage of the same passage type from a second document within the document execution environment to the user via a document passage comparison interface. Differences between the passages can be highlighted, enabling a user to quickly compare and contrast the passages.
1 . A computer-implemented method, comprising:
receiving a first document within a document execution environment, the first document to include multiple document passages;
applying a trained machine learning model to determine a level of risk associated with a first document passage of the first document based on at least in part on a passage type of the document passage, the trained machine learning model has been trained using a training set of information including training documents within the document execution environment, each training document including one or more document passages, the training set of information is separated into a positive training set and a negative training set, wherein the positive training set includes a subset of document passages, associated passage types, and associated risks designated as above a threshold level of risk, and the negative training set includes another subset of document passages, associated passage types and associated risks designated as below the threshold level of risk;
determining whether the level of risk associated with the first document passage exceeds a risk threshold by more than a threshold amount;
in response to determining that the level of risk associated with the first document passage exceeds the risk threshold, identifying a second document passage of the same passage type and related to the first document passage within a second document within the document execution environment, the second document passage not exceeding the risk threshold;
automatically performing a mitigating action associated with the first document passage in response to the determined level of risk exceeding the risk threshold by more than the threshold amount; and
presenting the first document passage on a first portion of a graphical user interface (GUI) of an electronic device and the second document passage on a second portion of the GUI of the electronic device.
2 . The method of claim 1 , wherein the trained machine learning model is trained on a set of training documents, each training document associated with the passage type, the level of risk, or a set of document characteristics, the set of document characteristics comprising a document type, a region, a language, or an industry.
3 . The method of claim 1 , further comprising presenting a GUI element to highlight a set of differences between the first document passage and the second document passage on the GUI of the electronic device.
4 . The method of claim 1 , further comprising:
generating a recommendation based on the level of risk associated with the first document passage by the trained machine learning model; and
presenting the recommendation on the GUI of the electronic device.
5 . The method of claim 1 , further comprising receiving feedback information indicating whether the determined level of risk associated with the first document passage of the first document is accurate or not accurate.
6 . The method of claim 1 , comprising modifying the training set of information for the trained machine learning model based on feedback information indicating whether the determined level of risk associated with the first document passage of the first document is accurate or not accurate.
7 . The method of claim 1 , further comprising re-training the trained machine learning model based on a modified training set of information.
8 . A non-transitory computer-readable storage medium storing instructions that, when executed by a hardware processor, cause the hardware processor to:
access a training set of information including training documents within a document execution environment, each training document including one or more passages, the training set of information is separated into a positive training set and a negative training set, wherein the positive training set includes a subset of document passages, associated passage types, and associated risks designated as above a threshold level of risk, and the negative training set includes another subset of document passages, associated passage types and associated risks designated as below the threshold level of risk;
train a machine learning model based on the accessed training set of information, the machine learning model is trained to determine a level of risk associated with a document passage based at least in part on the passage type of the document passage;
modify the training set of information based on a feedback received to indicate an accuracy of a level of risk determined by the machine learning model for one or more passages; and
re-train the machine learned model based on the modified training set of information.
9 . The non-transitory computer-readable storage medium of claim 8 , wherein the hardware processor is configured to present a recommendation to mitigate the determined level of risk in conjunction with the document passage and the determined level of risk.
10 . The non-transitory computer-readable storage medium of claim 8 , wherein each training document is associated with a set of document characteristics comprising a document type, a region, a language, or an industry, and wherein the machine learning model is trained to determine the level of risk associated with a document passage based additionally on the set of document characteristics associated with a document in which the document passage appears.
11 . The non-transitory computer-readable storage medium of claim 8 , wherein the passage type for a passage comprises at least one of: a type of a legal clause, a type of business clause, a type of finance clause, or a type of content within the passage.
12 . An apparatus, comprising:
a hardware processor; and
a non-transitory computer-readable storage medium storing executable instructions that, when executed by the hardware processor, causes the hardware processor to:
receive a first document within a document execution environment, the first document to include multiple document passages;
apply a trained machine learning model to determine a level of risk associated with a first document passage of the first document based on at least in part on a passage type of the document passage, the trained machine learning model has been trained using a training set of information including training documents within the document execution environment, each training document including one or more document passages, the training set of information is separated into a positive training set and a negative training set, wherein the positive training set includes a subset of document passages, associated passage types, and associated risks designated as above a threshold level of risk, and the negative training set includes another subset of document passages, associated passage types and associated risks designated as below the threshold level of risk;
determine whether the level of risk associated with the first document passage exceeds a risk threshold by more than a threshold amount;
in response to determining that the level of risk associated with the first document passage exceeds the risk threshold, identify a second document passage of the same passage type and related to the first document passage within a second document within the document execution environment, the second document passage not exceeding the risk threshold;
automatically perform a mitigating action associated with the first document passage in response to the determined level of risk exceeding the risk threshold by more than the threshold amount; and
present the first document passage on a first portion of a graphical user interface (GUI) of an electronic device and the second document passage on a second portion of the GUI of the electronic device.
13 . The apparatus of claim 12 , wherein the hardware processor is configured to present a GUI element to highlight a set of differences between the first document passage and the second document passage on the GUI of the electronic device.
14 . The apparatus of claim 12 , wherein the hardware processor is configured to:
generate a recommendation based on the level of risk associated with the first document passage by the trained machine learning model; and
present the recommendation on the GUI of the electronic device.
15 . The apparatus of claim 12 , wherein the passage type comprises at least one of: a type of a legal clause, a type of business clause, a type of finance clause, or a type of content within the passage.
16 . The apparatus of claim 12 , wherein the trained machine learning model is trained on a set of training documents, each training document associated with a passage type, the level of risk, or a set of document characteristics, the set of document characteristics comprising a document type, a region, a language, or an industry.