Methods and systems for facilitating family-based review
A active learning family-based review method includes selecting a document ranked as relevant by a machine learning model, identifying family documents relationally-linked to the ranked relevant document, generating a batch including the ranked relevant document adjacent to the family documents, and displaying the batch in a computing device. An active learning family-based review computing system includes a processor and a memory storing instructions that, when executed, cause the computing system to select a relevant document using machine learning, identify family documents, generate a batch including the relevant document adjacent to the family documents, and display the batch. A non-transitory computer readable medium stores program instructions that when executed, cause a computer system to select a relevant document using machine learning, identify family documents, generate a batch including the relevant document adjacent to the family documents, and display the batch.
1. A computer-implemented method for conducting family-based review of a set of documents in an active learning process, comprising:
selecting, from the set of documents in the active learning process, a document ranked as relevant by a machine learning model,
identifying a set of family documents relationally-linked to the ranked relevant document according to a hierarchical structure,
generating a batch of documents based on a relevance rank for the set of documents, wherein the relationally-linked family documents are included in the batch adjacent to the ranked relevant document, and
causing the batch of documents to be displayed on a display screen of a computing device.
2. The computer-implemented method of claim 1 , wherein selecting, from the set of documents in the active learning process, the document ranked as relevant by the machine learning model includes selecting from a set of documents with the highest relevance ranking certainty score.
3. The computer-implemented method of claim 1 , wherein selecting, from the set of documents in the active learning process, the document ranked as relevant by the machine learning model includes selecting from a set of documents with the lowest relevance ranking certainty score.
4. The computer-implemented method of claim 1 , wherein the set of documents comprise emails, and identifying the set of family documents relationally-linked to the ranked relevant document includes identifying one or more email attachments corresponding to the ranked relevant document.
5. The computer-implemented method of claim 1 , wherein identifying the set of family documents relationally-linked to the ranked relevant document includes clustering the set of documents using a clustering algorithm.
6. The computer-implemented method of claim 1 , wherein the relationally-linked family documents included in the batch adjacent to the ranked relevant document are represented in a tree data structure.
7. The computer-implemented method of claim 1 , further comprising
receiving, from a client computing device, a coding decision corresponding to a document, and
analyzing the coding decision and an identifier associated with the document to generate an updated machine learning model.
8. A family-based review computing system, comprising
one or more processors; and
a memory storing instructions that, when executed, cause the computing system to:
select, from a set of documents in an active learning process, a document ranked as relevant by a machine learning model,
identify a set of family documents relationally-linked to the ranked relevant document according to a hierarchical structure,
generate a batch of documents based on a relevance rank for the set of documents, wherein the relationally-linked family documents are included in the batch adjacent to the ranked relevant document, and
cause the batch of documents to be displayed on a display screen of a computing device.
9. The family-based review computing system of claim 8 , wherein the memory stores further instructions that when executed, cause the computing system to:
select from a set of documents with the highest relevance ranking certainty score.
10. The family-based review computing system of claim 8 , wherein the memory stores further instructions that when executed, cause the computing system to:
select from a set of documents with the lowest relevance ranking certainty score.
11. The family-based review computing system of claim 8 , wherein the memory stores further instructions that when executed, cause the computing system to:
identify one or more email attachments corresponding to the ranked relevant document.
12. The family-based review computing system of claim 8 , wherein the memory stores further instructions that when executed, cause the computing system to:
cluster the set of documents using a clustering algorithm.
13. The family-based review computing system of claim 8 , wherein the memory stores further instructions that when executed, cause the computing system to:
generate a tree data structure, wherein the ranked relevant document and relationally-linked family documents are adjacent in the tree data structure.
14. The family-based review computing system of claim 8 , wherein the memory stores further instructions that when executed, cause the computing system to:
receive, from a client computing device, a coding decision corresponding to a document, and
analyze the coding decision and an identifier associated with the document to generate an updated machine learning model.
15. A non-transitory computer readable medium storing program instructions that when executed, cause a computer system to:
select, from a set of documents in an active learning process, a document ranked as relevant by a machine learning model,
identify a set of family documents relationally-linked to the ranked relevant document according to a hierarchical structure,
generate a batch of documents based on a relevance rank for the set of documents, wherein the relationally-linked family documents are included in the batch adjacent to the ranked relevant document, and
cause the batch of documents to be displayed on a display screen of a computing device.
16. The non-transitory computer readable medium of claim 15 , including further program instructions that when executed, cause a computer system to:
select from a set of documents with the highest relevance ranking certainty score.
17. The non-transitory computer readable medium of claim 15 , including further program instructions that when executed, cause a computer system to:
select from a set of documents with the lowest relevance ranking certainty score.
18. The non-transitory computer readable medium of claim 15 , including further program instructions that when executed, cause a computer system to:
identify one or more email attachments corresponding to the ranked relevant document.
19. The non-transitory computer readable medium of claim 15 , including further program instructions that when executed, cause a computer system to:
cluster the set of documents using a clustering algorithm.
20. The non-transitory computer readable medium of claim 15 , including further program instructions that when executed, cause a computer system to: generate a tree data structure, wherein the ranked relevant document and relationally-linked family documents are adjacent in the tree data structure.