With　the　rising　incidences　of　emergencies,　it　is　both　challenging　and　meaningful　to　study　how　to　make　decisions　quickly　and　take　appropriate　measures　to　control　the　spread　and　evolution　of　the　situation.　However,　most　current　emergency　decision-making　focuses　on　mathematical　model　construction,　whereas　fuzzy　decision-making　is　biased　towards　subjective　assumptions,　which　are　both　insufficient　for　practicability.　We　studied　the　intelligent　acquisition　of　single-valued　neutrosophic　numbers　based　on　emotional　tendency　analysis　and　applied　them　to　emergency　decision-making.　First,　Python　programming　technology　was　used　to　crawl,　preprocess,　and　statistically　analyse　the　network　comment　data　of　emergencies,　and　a　quantised　single-value　neutrosophic　number　was　obtained.　Second,　the　attribute　values,　represented　as　neutrosophic　numbers,　were　uniformly　converted　into　cloud　droplets,　and　the　weight　of　the　attribute　values　was　objectively　determined　according　to　the　digital　feature　entropy　in　the　cloud　droplets.　Subsequently,　a　case-based　reasoning　approach　was　used　to　calculate　the　combined　weighted　similarity　between　the　alternatives　and　ideal　solution　(target　case)　to　obtain　a　ranking　of　the　alternatives　and　historical　cases　in　the　case　base　that　best　matches　the　target　case.　Finally,　a　typhoon　disaster　assessment　was　considered　as　an　example　to　verify　the　feasibility　and　effectiveness　of　the　proposed　method,　and　the　advantages　of　the　proposed　method　were　emphasised　through　multi-aspect　and　multi-angle　comparative　analyses.　The　relevant　research　can　be　used　for　public　opinion　monitoring　during　emergencies　and　emergency　handling.　　©　2023　-　IOS　Press.　All　rights　reserved.