Modeling　interpersonal　influence　on　different　sentiments　is　a　key　issue　for　opinion　formation　and　viral　marketing.　Previous　works　directly　define　interpersonal　influence　on　each　pair　of　users.　They　fail　to　depict　the　unobserved　relationships　between　user　pairs　and　thus　suffer　from　the　overfitting　problem　of　learning　users＇　influences.　Moreover,　there　are　still　not　effective　solutions　to　integrate　users＇　sentiments　to　understand　the　interpersonal　influence.　Therefore,　we　propose　a　user＇s　distributed　representation　model　with　sentimental　factors.　Firstly,　two　low-dimensional　parameter　matrices　are　applied　to　represent　opinion　propagators＇　influences　and　opinion　recipients＇　susceptibility　on　different　sentiments.　And　then,　we　describe　cascade　behaviors　with　the　survival　analysis　model.　Finally,　the　imbalance　of　positive　and　negative　cases　is　solved　by　employing　negative　case　sampling　technique,　according　to　the　distribution　of　infected　users＇　frequency.　Experimental　results　conducted　on　Microblog　database　with　different　sentiments　showed　that,　compared　to　the　state-of-the-art　models,　our　model　improved　273%　and　32.4%　on　MRR　metrics　on　＂Predicting　Cascade　Dynamics＂　and　＂Who　will　Be　Retweeted＂　tasks　respectively,　and　reduced　10.46%　on　MAPE　metrics　on　＂Cascade　Size　Predicting＂　task,　which　verified　the　validity　of　our　model.　Besides,　analyzing　the　distribution　of　learned　users＇　sentimental　influences　and　susceptibilities　resulted　in　some　important　discoveries.　©　2017,　Science　Press.　All　right　reserved.