In its raw frequency variety, tf is just the frequency of your "this" for each document. In Each individual document, the word "this" appears after; but given that the document 2 has a lot more terms, its relative frequency is scaled-down.log ⁡ N n t = − log ⁡ n t N displaystyle log frac N n_ t =-log frac n_ t N Once you included the necessar