Consumer　behavior　has　changed　due　to　digitization.　Online　shoppers　now　refer　to　user　reviews　containing　comprehensive　data　produced　in　real-time,　which　can　be　used　to　determine　users’　needs.　This　paper　combines　Kansei　engineering　and　natural　language　processing　techniques　to　extract　information　on　users’　needs　from　online　reviews　and　provide　guidance　for　subsequent　product　improvements　and　development.　A　crawler　tool　was　used　to　collect　a　large　number　of　online　reviews　for　a　target　product.　Frequency　analysis　was　then　applied　to　the　text　to　filter　out　the　product　components　worth　analyzing.　The　results　were　categorized　and　aggregated　by　experts　before　sentiment　analysis　was　performed　on　statements　containing　the　selected　adjectives.　Finally,　the　user　needs　identified　could　be　inputted　to　Kansei　engineering　for　further　product　design.　This　paper　verifies　the　merit　of　the　above　method　when　applied　to　the　mountain　bike　product　category　on　Amazon.　The　method　proved　to　be　a　quick　and　efficient　way　to　attain　accurate　product　evaluations　from　end-users　and　thus　represents　a　feasible　approach　to　intelligently　determining　user　preferences.　©　The　Author(s),　under　exclusive　licence　to　Springer-Verlag　London　Ltd.,　part　of　Springer　Nature　2024.