INTERACTIVE RECOMMENDATION OF DATA SETS FOR DATA ANALYSIS
A data analysis platform provides recommendations for datasets for analysis. Given a user selected dataset, for example resulting from a search, automatically identifies other datasets based a variety of different types of relationships, including lineage, structural, content, usage, classification, and organizational/social. Datasets for each type of relationship are identified and scored for relevance, and ranked. Selected ones of the ranked data sets are presented in a recommendation interface. As the user selects from recommended dataset, additional datasets are automatically recommended based in inferences made according to the selected dataset and relationship.
1 . A computer executed method of recommending datasets for data analysis, comprising:
receiving a user selection of a first dataset;
determining a context corresponding to the user selection of the first dataset;
determining, based on the first dataset and determined context, one or more dataset recommenders, each of the one or more recommenders corresponding to a relationship type between datasets;
determining a plurality of second datasets related to the first dataset based on the relationship types;
scoring each of the plurality of second datasets using a relevance ranking algorithm specific to the corresponding relationship type to score the relevance of the of the second dataset to first dataset;
ranking the plurality of second datasets based on the scoring;
selecting a subset of the ranked datasets as the recommended datasets; and
presenting the recommended datasets in a graphical user interface, wherein the recommended datasets are grouped by relationship type to the first dataset.
2 . The computer executed method of claim 1 , wherein the relationship types comprise relationship types selected from the group consisting of:
a lineage relationship based on ancestor or descendant relationships between datasets;
a content relationship based on semantically similar datasets;
a structure relationship based on structurally compatible datasets;
a usage based relationships based on datasets previously used by relevant classes of users in association with the previously chosen datasets;
a classification-based relationship based on datasets that share one or more classifications with one or more datasets previously chosen by the user; and;
an organizational or social relationship based on social or organizational relationships between users of the datasets.
3 . The computer executed method of claim 1 , further comprising:
in response to receiving a selection of one or more recommended datasets, providing a second level of recommended datasets, comprising:
determining a second context corresponding to the user selection of the one or more recommended datasets;
determining, based on the one or more recommended datasets and determined second context, one or more dataset recommenders;
determining a plurality of third datasets related to the one or more recommended datasets based on the relationship types;
scoring each of the plurality of third datasets using the relevance ranking algorithm;
ranking the plurality of third datasets based on the scoring;
selecting a subset of the ranked third datasets as the second level of recommended datasets; and
presenting the second level of recommended datasets in the graphical user interface, wherein the second level of recommended datasets are grouped by relationship type to the selected dataset.
4 . The computer executed method of claim 1 , further comprising:
in response to determining the context corresponding to the user selection of the first dataset, inferring a user goal based on the context for the user selection of the first dataset; and
presenting the inferred user goal in the a graphical user interface.
5 . The computer executed method of claim 4 , further comprising:
receiving user input adjusting the inferred user goal presented in the a graphical user interface to a replacement goal;
in response to the adjusting:
determining a revised plurality of datasets related to the first dataset based on the replacement goal;
scoring each of the revised plurality of datasets using a relevance ranking algorithm specific to the corresponding relationship type to score the relevance of the of the second dataset to first dataset;
ranking the revised plurality of datasets based on the scoring;
selecting a revised subset of the ranked datasets as a revised set of recommended datasets; and
replacing the recommended datasets in the graphical user interface with the revised set of recommended datasets.
6 . The computer executed method of claim 4 , further comprising:
receiving user input adjusting the inferred user goal presented in the graphical user comprising rejection of the presented inferred goal.
7 . The computer executed method of claim 4 , wherein the inferred user goal is based on a class associated with the determined context and actions associated with the class.
8 . The computer executed method of claim 4 , wherein the inferred user goal is selected from the group consisting of finding a cleaner dataset, enriching the dataset, and integrating datasets.
9 . The computer executed method of claim 1 , wherein scoring each of the plurality of second datasets further comprises:
within each relationship type, scoring the second datasets of the relationship type by relevance to the first dataset; and
wherein ranking the plurality of second datasets based on the scoring is based on the scoring within each relationship type and a further scoring of the relationship types.
10 . The computer executed method of claim 1 , further comprising:
generating a preview of contents of a recommended dataset of the presented recommended datasets in the graphical user interface; and
in response to user input selecting the recommended dataset, presenting the preview of the recommended dataset to the user in the graphical user interface.
11 . A non-transitory computer-readable memory storing a computer program executable by a processor, the computer program producing a user interface displaying dataset recommendations, the user interface comprising:
a dataset selection control for receiving a user selection of a first dataset;
a recommendation bar for presenting a set of recommended datasets based on the user selection of the first dataset and a determined context for the selection, wherein the recommended datasets are grouped within the recommendation bar by relationship type to the first dataset;
a relationship confirmation control for receiving a selection of one or more of the recommended datasets.
12 . The computer program of claim 11 , wherein the user interface is further configured by the computer program to:
in response to receiving a selection of one or more of the recommended datasets, presenting a second level of recommended datasets in the graphical user interface, wherein the second level of recommended datasets are grouped by relationship type to the selected dataset.
13 . The computer program of claim 11 , further comprising:
presenting an inferred user goal in the a graphical user interface, the inferred user goal based on the determined context for the user selection of the first dataset.
14 . The computer program of claim 13 , further comprising:
in response to receiving user input adjusting the inferred user goal presented in the graphical user interface to a replacement goal, replacing the recommended datasets in the graphical user interface with a revised set of recommended datasets.
15 . The computer program of claim 13 , further comprising:
in response to receiving user input adjusting the inferred user goal presented in the graphical user interface comprising rejection of the presented inferred goal, replacing the recommended datasets in the graphical user interface with a revised set of recommended datasets.
16 . The computer program of claim 11 , further comprising:
in response to user input selecting the recommended dataset, presenting a preview of the recommended dataset to the user in the graphical user interface.
17 . A computer program product comprising a non-transitory computer readable storage medium having instructions encoded therein that, when executed by a processor, cause the processor to:
receiving a user selection of a first dataset;
determining a context corresponding to the user selection of the first dataset;
determining, based on the first dataset and determined context, one or more dataset recommenders, each of the one or more recommenders corresponding to a relationship type between datasets;
determining a plurality of second datasets related to the first dataset based on the relationship types;
scoring each of the plurality of second datasets using a relevance ranking algorithm specific to the corresponding relationship type to score the relevance of the of the second dataset to first dataset;
ranking the plurality of second datasets based on the scoring;
selecting a subset of the ranked datasets as the recommended datasets; and
presenting the recommended datasets in a graphical user interface, wherein the recommended datasets are grouped by relationship type to the first dataset.
18 . The computer program product of claim 17 , further comprising instructions encoded therein that, when executed by the processor, cause the processor to perform steps comprising:
in response to receiving a selection of one or more recommended datasets, providing a second level of recommended datasets, comprising:
determining a second context corresponding to the user selection of the one or more recommended datasets;
determining, based on the one or more recommended datasets and determined second context, one or more dataset recommenders;
determining a plurality of third datasets related to the one or more recommended datasets based on the relationship types;
scoring each of the plurality of third datasets using the relevance ranking algorithm;
ranking the plurality of third datasets based on the scoring;
selecting a subset of the ranked third datasets as the second level of recommended datasets; and
presenting the second level of recommended datasets in the graphical user interface, wherein the second level of recommended datasets are grouped by relationship type to the selected dataset.
19 . The computer program product of claim 17 , further comprising instructions encoded therein that, when executed by the processor, cause the processor to perform steps comprising:
in response to determining the context corresponding to the user selection of the first dataset, inferring a user goal based on the context for the user selection of the first dataset; and
presenting the inferred user goal in the a graphical user interface.
20 . The computer program product of claim 19 , further comprising instructions encoded therein that, when executed by the processor, cause the processor to perform steps comprising:
receiving user input adjusting the inferred user goal presented in the a graphical user interface to a replacement goal;
in response to the adjusting:
determining a revised plurality of datasets related to the first dataset based on the replacement goal;
scoring each of the revised plurality of datasets using a relevance ranking algorithm specific to the corresponding relationship type to score the relevance of the of the second dataset to first dataset;
ranking the revised plurality of datasets based on the scoring;
selecting a revised subset of the ranked datasets as a revised set of recommended datasets; and
replacing the recommended datasets in the graphical user interface with the revised set of recommended datasets.
21 . The computer program product of claim 17 , wherein scoring each of the plurality of second datasets further comprises:
within each relationship type, scoring the second datasets of the relationship type by relevance to the first dataset; and
wherein ranking the plurality of second datasets based on the scoring is based on the scoring within each relationship type and a further scoring of the relationship types.
22 . The computer program product of claim 17 , further comprising instructions encoded therein that, when executed by the processor, cause the processor to perform steps comprising:
generating a preview of contents of a recommended dataset of the presented recommended datasets in the graphical user interface; and
in response to user input selecting the recommended dataset, presenting the preview of the recommended dataset to the user in the graphical user interface.