Enabling rhetorical analysis via the use of communicative discourse trees
Systems, devices, and methods of the present invention calculate a rhetorical relationship between one or more sentences. In an example, a computer-implemented method accesses a sentence comprising a plurality of fragments. At least one fragment includes a verb and a words. Each word includes a role of the words within the fragment. Each fragment is an elementary discourse unit. The method generates a discourse tree that represents rhetorical relationships between the sentence fragments. The discourse tree includes nodes including nonterminal and terminal nodes, each nonterminal node representing a rhetorical relationship between two of the sentence fragments, and each terminal node of the nodes of the discourse tree is associated with one of the sentence fragments. The method matches each fragment that has a verb to a verb signature, thereby creating communicative discourse tree.
1. A computer-implemented method for determining a complementarity of a pair of two sentences by analyzing communicative discourse trees, the method comprising:
determining, for a question sentence, a question communicative discourse tree comprising a question root node;
determining, for an answer sentence, an answer communicative discourse tree, wherein the answer communicative discourse tree comprises an answer root node, wherein determining a communicative discourse tree from a discourse tree comprises matching, in the discourse tree, each elementary discourse unit that has a verb to a verb signature,
wherein each verb signature comprises the verb and a respective thematic role that describes a relationship between the verb and other words;
responsive to identifying that the question root node and the answer root node are identical, merging the communicative discourse trees;
creating a syntactic parse tree from the question sentence and the answer sentence;
merging the syntactic parse tree and the merged communicative discourse tree into a parse thicket;
computing a level of complementarity between the question communicative discourse tree and the answer communicative discourse tree by applying a predictive model to the parse thicket; and
responsive to determining that the level of complementarity is above a threshold, identifying the question sentence and the answer sentence as complementary and sending text associated with either the question sentence or the answer sentence to an external device.
2. The method of claim 1 , further comprising associating the question communicative discourse tree and the answer communicative discourse tree by creating an additional communicative discourse tree.
3. The method of claim 1 , wherein the predictive model is trained with a thicket learning machine learning algorithm to determine a level of complementarity of sub-trees of two communicative discourse trees, wherein the thicket machine learning algorithm involves training with one or more parse thickets.
4. The method of claim 1 , further comprising:
aligning an elementary discourse unit of the merged communicative discourse tree with the syntactic parse tree; and
merging the elementary discourse unit of the merged communicative discourse tree and the syntactic parse tree.
5. The method of claim 1 , wherein the predictive model is a machine learning algorithm that is trained to determine an additional level of complementarity between two communicative discourse trees identified as matching or not matching.
6. The method of claim 1 , wherein the merged communicative discourse tree is a first communicative discourse tree, further comprising:
identifying a best match between the first communicative discourse tree and a second communicative discourse tree from a plurality of communicative discourse trees; and
sending text associated with the second communicative discourse tree to a device.
7. The method of claim 1 , further comprising:
sending, to a device, (i) the question sentence or (ii) the answer sentence as a search result.
8. A non-transitory computer-readable medium storing computer-executable program instructions that when executed, configure a processing device to perform operations comprising:
determining, for a question sentence, a question communicative discourse tree comprising a question root node;
determining, for an answer sentence, an answer communicative discourse tree, wherein the answer communicative discourse tree comprises an answer root node, wherein determining a communicative discourse tree from a discourse tree comprises matching, in the discourse tree, each elementary discourse unit that has a verb to a verb signature,
wherein each verb signature comprises the verb and a respective thematic role that describes a relationship between the verb and other words;
responsive to identifying that the question root node and the answer root node are identical, merging the communicative discourse trees;
creating a syntactic parse tree from the question sentence and the answer sentence;
merging the syntactic parse tree and the merged communicative discourse tree into a parse thicket;
computing a level of complementarity between the question communicative discourse tree and the answer communicative discourse tree by applying a predictive model to the parse thicket; and
responsive to determining that the level of complementarity is above a threshold, identifying the question sentence and the answer sentence as complementary and sending text associated with either the question sentence or the answer sentence to an external device.
9. The non-transitory computer-readable medium of claim 8 , wherein when executed, the computer-executable program instructions configure a processing device to perform operations comprising:
accessing a set of training data comprising a set of training pairs, wherein each training pair comprises a question communicative discourse tree that represents a question and an answer communicative discourse tree that represents an answer and an expected level of complementarity; and
training the predictive model by iteratively:
providing one of the training pairs to the predictive model,
receiving, from the predictive model, a determined level of complementarity;
adjusting internal parameters of the predictive model to reduce an error, wherein the error is a difference between the determined level of complementarity and the respective expected level of complementarity;
using the predictive model to identify a correspondence between a first sentence and a second sentence; and
providing either the first sentence or the second sentence to a device.
10. The non-transitory computer-readable medium of claim 9 , wherein the set of training data comprises a question answer pair that comprises (i) a question and (ii) an answer that is relevant but is rhetorically incorrect when compared to the question.
11. The non-transitory computer-readable medium of claim 9 , wherein when executed, the computer-executable program instructions configure a processing device to perform operations comprising:
creating a merged communicative discourse tree from a question communicative discourse tree representing a question and an answer communicative discourse tree representing an answer; and
computing an additional level of complementarity between the question communicative discourse tree and the answer communicative discourse tree by applying the predictive model to the merged communicative discourse tree.
12. The non-transitory computer-readable medium of claim 9 , wherein when executed, the computer-executable program instructions configure a processing device to perform operations comprising associating the question communicative discourse tree and the answer communicative discourse tree by creating an additional communicative discourse tree.
13. The non-transitory computer-readable medium of claim 9 , wherein the predictive model is trained with a thicket learning machine learning algorithm to determine a level of complementarity of sub-trees of two communicative discourse trees, wherein the thicket machine learning algorithm involves training with one or more parse thickets.
14. A system comprising:
a non-transitory computer-readable medium storing computer-executable program instructions; and
a processing device communicatively coupled to the non-transitory computer-readable medium for executing the computer-executable program instructions, wherein executing the computer-executable program instructions configures the processing device to perform operations comprising:
determining, for a question sentence, a question communicative discourse tree comprising a question root node;
determining, for an answer sentence, an answer communicative discourse tree, wherein the answer communicative discourse tree comprises an answer root node, wherein determining a communicative discourse tree from a discourse tree comprises matching, in the discourse tree, each elementary discourse unit that has a verb to a verb signature,
wherein each verb signature comprises the verb and a respective thematic role that describes a relationship between the verb and other words;
responsive to identifying that the question root node and the answer root node are identical, merging the communicative discourse trees;
creating a syntactic parse tree from the question sentence and the answer sentence;
merging the syntactic parse tree and the merged communicative discourse tree into a parse thicket;
computing a level of complementarity between the question communicative discourse tree and the answer communicative discourse tree by applying a predictive model to the parse thicket; and
responsive to determining that the level of complementarity is above a threshold, identifying the question sentence and the answer sentence as complementary and sending text associated with either the question sentence or the answer sentence to an external device.
15. The system of claim 14 , wherein executing the computer-executable program instructions configures the processing device to perform operations comprising:
aligning an elementary discourse unit of the merged communicative discourse tree with a corresponding syntactic parse tree; and
merging the elementary discourse unit of the merged communicative discourse tree and the corresponding syntactic parse tree.
16. The system of claim 14 , wherein the predictive model is a machine learning algorithm that is trained to determine an additional level of complementarity between two communicative discourse trees identified as matching or not matching.
17. The system of claim 14 , wherein executing the computer-executable program instructions configures the processing device to perform operations comprising:
sending, to a device, one or both of (i) the question sentence or (ii) the answer sentence as a search result.
18. The system of claim 14 , wherein executing the computer-executable program instructions configures the processing device to perform operations comprising:
accessing a set of training data comprising a set of training pairs, wherein each training pair comprises a question communicative discourse tree that represents a question and an answer communicative discourse tree that represents an answer and an expected level of complementarity; and
training the predictive model by iteratively:
providing one of the training pairs to the predictive model,
receiving, from the predictive model, a determined level of complementarity; and
adjusting internal parameters of the predictive model to reduce an error, wherein the error is a difference between the determined level of complementarity and the respective expected level of complementarity.
19. The system of claim 18 , wherein the set of training data comprises a question answer pair that comprises a question and an answer that is relevant but is rhetorically incorrect when compared to the question.
20. The system of claim 14 , wherein executing the computer-executable program instructions configures the processing device to perform operations comprising:
creating a merged communicative discourse tree from a question communicative discourse tree representing a question and an answer communicative discourse tree representing an answer; and
computing an additional level of complementarity between the question communicative discourse tree and the answer communicative discourse tree by applying the predictive model to the merged communicative discourse tree.