A Flexible Table Parsing Approach
Fuente:
Redalyc
Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Artículo científico |
| Lingua: | en |
| Pubblicazione: |
Instituto Politécnico Nacional
2012
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1876472986141196288 |
|---|---|
| author | Frank Schilder |
| author_facet | Frank Schilder |
| contents | A Flexible Table Parsing Approach Frank Schilder Ravi Kondadadi Yana Kadiyska Computación tables document processing Information retrieval Relational data is often encoded in tables. Tables areeasy to read by humans, but difficult to interpret automatically.In cases where table layout cues are not obtainable (missingHTML tags) or where columns are distorted (by copying from aspreadsheet to text) previous table extraction approaches run intoproblems. This paper introduces a novel table parsing approach.Our approach is based on a set of simple assumptions: (a) everytable can be split up in data cells and headers, and (b) every tablecan be parsed beginning from a data cell utilizing the overall tablestructure. The table parsing is defined as “table flattening” in thispaper. That is, the parsing starts with a data cell and pulls out alltoken (i.e., headers and sub-headers) associated with a respectivedata cell. We propose a parsing technique that uses two simpleparsing heuristics: table headers are to the left of and above adata cell. We experimented with trader emails that containedinstrument information with bid-ask prices as data cells. Wedeveloped a clustering and classifying method for finding pricesreliably in the data set we used. This method is transferable toother data cell types and can be applied to other table content. 2012 artículo científico 1870-9044 https://www.redalyc.org/articulo.oa?id=402640459003 en http://www.redalyc.org/revista.oa?id=4026 Polibits application/pdf Instituto Politécnico Nacional Polibits (México) Vol.45 |
| format | Artículo científico |
| id | redalyc_402640459003 |
| institution | Redalyc |
| language | en |
| publishDate | 2012 |
| publisher | Instituto Politécnico Nacional |
| spellingShingle | A Flexible Table Parsing Approach Frank Schilder Computación tables document processing Information retrieval A Flexible Table Parsing Approach Frank Schilder Ravi Kondadadi Yana Kadiyska Computación tables document processing Information retrieval Relational data is often encoded in tables. Tables areeasy to read by humans, but difficult to interpret automatically.In cases where table layout cues are not obtainable (missingHTML tags) or where columns are distorted (by copying from aspreadsheet to text) previous table extraction approaches run intoproblems. This paper introduces a novel table parsing approach.Our approach is based on a set of simple assumptions: (a) everytable can be split up in data cells and headers, and (b) every tablecan be parsed beginning from a data cell utilizing the overall tablestructure. The table parsing is defined as “table flattening” in thispaper. That is, the parsing starts with a data cell and pulls out alltoken (i.e., headers and sub-headers) associated with a respectivedata cell. We propose a parsing technique that uses two simpleparsing heuristics: table headers are to the left of and above adata cell. We experimented with trader emails that containedinstrument information with bid-ask prices as data cells. Wedeveloped a clustering and classifying method for finding pricesreliably in the data set we used. This method is transferable toother data cell types and can be applied to other table content. 2012 artículo científico 1870-9044 https://www.redalyc.org/articulo.oa?id=402640459003 en http://www.redalyc.org/revista.oa?id=4026 Polibits application/pdf Instituto Politécnico Nacional Polibits (México) Vol.45 |
| title | A Flexible Table Parsing Approach |
| topic | Computación tables document processing Information retrieval |
| url | https://www.redalyc.org/articulo.oa?id=402640459003 |