USER CUSTOMISED SEARCH ENGINE USING MACHINE LEARNING, NATURAL LANGUAGE PROCESSING AND READABILITY ANALYSIS

摘要

A computer implemented method, computerised system and computer readable recording medium are disclosed. In one aspect, the computer implemented method comprises steps of: generating an index of electronic objects, comprising steps of: performing, upon each electronic object, natural language processing, using a natural language processor, and readability analysis, using a readability analyser, to generate natural language processing data and a readability score for the respective object; and storing in the index, for each electronic object, electronic object data indicative of the electronic object and associated meta-data indicative of the natural language processing data and the readability score; receiving, from a user operating a user processing system, a request to perform a search of the index, wherein the request has associated therewith input data; performing, upon the input data, natural language processing, using the natural language processor to generate natural language processing data and a readability range for the input data; searching, using a search engine, the index based upon the readability range and the natural language processing data for the input data, to determine a plurality of correlated electronic objects; generating, using a ranking engine, a ranked list of the plurality of correlated electronic objects based on a correlation level for each of the plurality of electronic objects and a machine learned rating generated by a machine learned rating prediction system trained according to a profile associated with the user, the profile being indicative of natural language processing data generated, by the natural language processor, based on previous search requests received from the user and performed by the search engine; and transferring, to the user processing system, search results indicative of at least a portion of the ranked list of the plurality of correlated electronic objects.