Is it possible to recognize in FineReader only number to improve the OCR quality for the file that contains only numbers? E.g. “5” sometimes reads as an “S” or a “1” reads as an “I”.
To recognize only numbers try to select the recognition language Numbers (available under the Formal languages) in Tools > Options:
Also, you can define language (e.g. Numbers) for the specified area in the Area Properties panel of OCR Editor: