2021 Fiscal Year Research-status Report
Development of Fast and Highly Effective Feature Subset Selection Algorithms based on Novel Integration of Quantum Computing and Machine Learning
| Project/Area Number |
20K11939
|
|---|
| Research Institution | Iwate Prefectural University |
|---|
Principal Investigator |
チャクラボルティ バサビ 岩手県立大学, ソフトウェア情報学部, 教授 (90305293)
|
|---|
| Project Period (FY) |
2020-04-01 – 2023-03-31
|
| Keywords | feature subset selection / metaheuristic algorithm / quantum inspired |
|---|
| Outline of Annual Research Achievements |
This year we have extended and applied our previously developed quantum inspired owl search algorithm (QIOSA) in last year for high dimensional microarray gene data. The algorithm has been modified to two step process to reduce computational time for high dimensional data by incorporating an ensemble based filter feature selection in the first step and a filter based search function in the final step to find out the optimal feature subset. Simulation experiments with several high dimensional data show that the proposed algorithm performs better than many state-of-the-art similar feature selection algorithms in terms of cardinality of the selected feature subset, classification accuracy and computational time.
|
| Current Status of Research Progress |
Current Status of Research Progress
3: Progress in research has been slightly delayed.
Reason
Due to COVID situations in India, we could not collaborate properly with our research collaborators in India. We could not execute our previous plans as institutions were closed for a long time. We could not also visit India to have longer collaborations for implementation of our algorithms and having input for one of the collaborators regarding quantum computing. We only could collaborate via zoom meetings.
|
| Strategy for Future Research Activity |
As now COVID situation is slightly improved, principal investigator is planning to visit Collaborator's university to expedite the research, complete the implementation part pf quantum feature selection algorithm and proposal of the general framework of quantum machine learning with the help of members of research collaborators.
|
| Causes of Carryover |
We could not attend international conference and use foreign resources due to border restrictions and unavailability respectively.
This year we plan to 1) Attend International conference to present our research result 2) Use foreign resources (quantum computer platform) for completion of research 3) Publish our research results in international journal
|
