| A00-2024 | human-written abstracts , we developed a | decomposition | program . The automatic decomposition |
| A00-2024 | Baum , 1972 ) solution to the | decomposition | problem . We first mathematically |
| A00-1008 | means-end reasoning . Hierarchical | decomposition | is more appropriate to dialogue |
| A00-2024 | A human subject then read the | decomposition | results of these sentences to |
| A00-1030 | with the s . In such cases , the | decomposition | using the singular first word |
| A00-2024 | summary alignment task . We ran the | decomposition | program to identify the source |
| A00-2024 | its most likely location . This | decomposition | program allows us to analyze |
| A00-1008 | and Nau 1994 ) . Hierarchical | decomposition | asserts that each goal can be |
| A00-2024 | This corpus was created using the | decomposition | program . We compute three types |
| A00-1008 | planning . Second , Hierarchical | decomposition | minimizes search time . Third |
| A00-2024 | The main focus of our work is on | decomposition | of summaries , sentence reduction |
| A00-2024 | overall system . We evaluated the | decomposition | program by two experiments , |
| A00-2024 | human-written abstracts . The results from | decomposition | are used to build the training |
| A00-1032 | character is a MF . Figure 4 shows | decomposition | of sentences into MFs ( enclosed |
| A00-1008 | explicitly specified . Hierarchical | decomposition | provides this trace naturally |
| A00-1008 | aware of the context in which a | decomposition | is proposed . Items in the hiercx |
| A00-2024 | decomposition program . The automatic | decomposition | allows us to build large corpora |
| A00-2024 | written from scratch . We used our | decomposition | program to automatically analyze |
| A00-2024 | reduction , sentence com - bination , | decomposition | , and sentence selection , are |
| A00-2024 | human-written summary sentences The | decomposition | program , see ( Jing and McKe |
