hello new to this forum working on a tool to extract data from different type of files such as doc,txt,pdf.(using jsp) Data in files may be unstructed, it is a bioliogical data want to extract scientific name ,author ,locality etc fields.while extracting want to make the application intelligent suppose below is data in file
DN: Database Name ASFA: Aquatic Sciences and Fisheries Abstracts, 1971-1993 TI: Title Introduction: Review of coastal erosion problems in the western Indian Ocean Region AU: Author Granlund, A CA: Corporate Author IOC, Paris (France)
--- first time the user has to move field name & it's value to database but next time when same type of file comes it should go to database automatically. that is want to make the program as a learning system. the program should learn abt the file structure & when again same type of file comes should do it automatically
for this what should i store abt th i/p file patterns? what techniques & how can be used such as artificial intelligence or neural networks or any other? if any one has does this before plz let me know
My suggestion is that you should use Excel file sheets or .csv files, coz these types of files are structured. If you are restricted to diff file patterns and that too if .txt or .doc types then you should use Proper heading and text seperated with some particular delimiter.
but the input data which i gave in my question was just one case
so whenever a file comes i want to store the pattern
so how to store the pattern & use it next time when same type of file comes
i.e what technique should be used for storing the pattern of file ----- DN: Database Name ASFA: Aquatic Sciences and Fisheries Abstracts, 1971-1993 TI: Title Introduction: Review of coastal erosion problems in the western Indian Ocean Region AU: Author Granlund, A CA: Corporate Author IOC, Paris (France)
such as i want to store for above file that there are DN, TI, AU, CA fields & next to them are their values
Previous Topic
