Context-sensitive Natural Language Processing with Memory Mechanism
| Project/Area Number |
18K11475
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Multi-year Fund |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 61040:Soft computing-related
|
|---|
| Research Institution | Osaka Prefecture University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2018-04-01 – 2022-03-31
|
| Project Status |
Completed (Fiscal Year 2021)
|
| Budget Amount *help |
¥4,420,000 (Direct Cost: ¥3,400,000、Indirect Cost: ¥1,020,000)
Fiscal Year 2021: ¥520,000 (Direct Cost: ¥400,000、Indirect Cost: ¥120,000)
Fiscal Year 2020: ¥520,000 (Direct Cost: ¥400,000、Indirect Cost: ¥120,000)
Fiscal Year 2019: ¥910,000 (Direct Cost: ¥700,000、Indirect Cost: ¥210,000)
Fiscal Year 2018: ¥2,470,000 (Direct Cost: ¥1,900,000、Indirect Cost: ¥570,000)
|
|---|
| Keywords | 自然言語処理 / 機械学習 / 深層学習 / アテンション / 対話生成 |
|---|
| Outline of Final Research Achievements |
Deep Learning contributes to improving natural language processing. Especially, in machine translation, it improves translation accuracy dramatically. Based on these results, we need to accelerate implementation of intelligent systems for social services. Especially, it is important to communicate a computer with more natural language expression.
In this research, we realize language generation according to preceding contexts and try to construct more flexible natural language processing. Speaking concretely, we are able to construct a neural conversation system that changes reply utterances according to the preceding utterances.
|
| Academic Significance and Societal Importance of the Research Achievements |
学術的意義としては、深層学習による言語生成に対して、外部からの制御可能性を示すことができ、深層学習システムのブラックボックス化を解消する糸口を提示することができた。
社会的意義としては、先行する発話に応じて人工知能システムからの応答が変わることにより、計算機と協働して作業を行えることを示すことができたため、システムに対する不透明性を解消することで、社会への人工知能成果の還元に貢献するものである。
|
Report
(5 results)
Research Products
(14 results)
