MARKET BASKET ANALYSIS FOR INFANT HYBRID TECHNOLOGY DETECTION
One example method includes identifying emerging hybrid technologies. Datasets are data mined to identify transactions that are related to technology. Associations are generated based on the transactions and the associations are processed to identify emerging hybrid technologies from the transactions.
1 . A method, comprising:
receiving transactions, wherein each of the transactions are associated with one or more items and wherein each of the transactions is associated with a technology;
generating metrics from the transactions;
determining associations from the metrics, wherein the associations identify hybrid technologies; and
outputting predictions, the predictions including at least some of the hybrid technologies, wherein each of the hybrid technologies is associated with at least two of the transactions.
2 . The method of claim 1 , further comprising receiving, as input, a dataset that includes the transactions.
3 . The method of claim 1 , further comprising:
receiving, as input, a dataset;
processing the dataset to extract terms, wherein each term is associated with one of the technologies; and
classifying the extracted terms such that the terms are unified across the dataset.
4 . The method of claim 3 , wherein the dataset includes a plurality of different datasets.
5 . The method of claim 3 , further comprising determining a phase for each of the transactions, wherein each phase is one of a genesis phase, a custom-built phase, a product phase, or a commodity phase.
6 . The method of claim 5 , further comprising filtering the transactions that are in the product phase or the commodity phase into filtered transactions.
7 . The method of claim 6 , wherein the metrics are generated from the filtered transactions.
8 . The method of claim 7 , wherein the metrics are generated by generating a lift metric, a confidence metric, and a support metric, wherein the extracted transactions are above at least one of a lift threshold, a confidence threshold, and/or a support threshold.
9 . The method of claim 1 , further comprising processing the predictions by merging at least some of the associations using one or more models.
10 . The method of claim 9 , wherein the models include one or more of a splitting compounded technology model and a word to vector model.
11 . A non-transitory storage medium having stored therein instructions that are executable by one or more hardware processors to perform operations comprising:
receiving a transactions, wherein each of the transactions are associated with one or more items and wherein each of the transactions is associated with a technology;
generating metrics from the transactions;
determining associations from the metrics, wherein the associations identify hybrid technologies; and
outputting predictions, the predictions including at least some of the hybrid technologies, wherein each of the hybrid technologies is associated with at least two of the transactions.
12 . The non-transitory storage medium of claim 11 , further comprising receiving, as input, a dataset that includes the transactions.
13 . The non-transitory storage medium of claim 11 , further comprising:
receiving, as input, a dataset;
processing the dataset to extract terms, wherein each term is associated with one of the technologies; and
classifying the extracted terms such that the terms are unified across the dataset.
14 . The non-transitory storage medium of claim 13 , wherein the dataset includes a plurality of different datasets.
15 . The non-transitory storage medium of claim 13 , further comprising determining a phase for each of the transactions, wherein each phase is one of a genesis phase, a custom-built phase, a product phase, or a commodity phase.
16 . The non-transitory storage medium of claim 15 , further comprising filtering the transactions that are in the product phase or the commodity phase into filtered transactions.
17 . The non-transitory storage medium of claim 16 , wherein the metrics are generated from the filtered transactions.
18 . The non-transitory storage medium of claim 17 , wherein the metrics are generated by generating a lift metric, a confidence metric, and a support metric, wherein the extracted transactions are above at least one of a lift threshold, a confidence threshold, and/or a support threshold.
19 . The non-transitory storage medium of claim 11 , further comprising processing the predictions by merging at least some of the associations using one or more models.
20 . The non-transitory storage medium of claim 19 , wherein the models include one or more of a splitting compounded technology model and a word to vector model.