Automated digital text optimization and modification

An automated digital text optimization method comprising:

• receiving, by a processor of a recurrent neural network (RNN) encoder-decoder hardware device during execution of an Internet search process, digital textual content comprising embedded words, morphemes, and phrases;
• identifying, by the processor via executing token appending code, a specified format for generating a summary for the digital textual content, the specified format indicating a word of interest within the digital textual content that is to be retained in the summary for the digital textual content;
• prepending a digital start token to a beginning portion of the word of interest and append a digital end token to an ending portion of the word of interest to obtain modified digital textual content;
• generating, by the processor via a decoder component of the RNN encoder-decoder hardware device with respect to the specified format, probability distribution attributes with respect to target vocabulary of the digital textual content;
• replacing, by the processor in response to analysis of the probability distribution attributes, a predicted word of the digital textual content of a time step with a word of target vocabulary with a highest probability distribution attribute of the probability distribution attributes;
• generating, by the processor, based at least in part on replacing the predicted word, an automated digital summary of the digital textual content that includes the word of interest, thereby enabling optimized operational functionality of the RNN encoder-decoder hardware device;
• and
• presenting, by the processor to a user via a specialized graphical user interface (GUI), a decoded version of the automated digital summary.

A non-transitory computer-readable medium storing a set of instructions for automated digital text optimization, the set of instructions comprising:

• one or more instructions that, when executed by one or more processors of a device, cause the device to:
• receive, during execution of an Internet search process, digital textual content comprising embedded words, morphemes, and phrases;
• identify, via executing token appending code, a specified format for generating a summary for the digital textual content, the specified format indicating a word of interest within the digital textual content that is to be retained in the summary for the digital textual content to obtain modified digital textual content;
• prepend a digital start token to a beginning portion of the word of interest and append a digital end token to an ending portion of the word of interest;
• generate, via a decoder component of a RNN encoder-decoder hardware device and based at least in part on a context vector corresponding to the modified digital textual content, probability distribution attributes with respect to target vocabulary of the digital textual content;
• replace, in response to analysis of the probability distribution attributes, a predicted word of the digital textual content of a time step with a word of target vocabulary with a highest probability distribution attribute of the probability distribution attributes;
• generate, based at least in part on replacing the predicted word, an automated digital summary of the digital textual content that includes the word of interest, thereby enabling optimized operational functionality of the RNN encoder-decoder hardware device;
• and
• present, to a user via a specialized graphical user interface (GUI), a decoded version of the automated digital summary.