Sintered　nano-silver　is　a　widely　used　interconnect　material　in　the　third　generation　of　power　semiconductors,　and　is　often　used　for　service　in　extremely　high-temperature　environments　due　to　its　advantages　of　high　melting　temperature,　making　the　reliability　of　interconnection　more　and　more　important　for　the　overall　reliability　of　the　third-generation　power　semiconductor　device.　Different　from　traditional　fatigue　life　prediction　methods,　a　thermal　fatigue　lifetime　dataset　is　collected　based　on　Coffin-Manson　thermal　fatigue　model　and　finite　element　simulations　in　this　paper.　A　data-driven　method　based　on　the　Support　Vector　Regression　(SVR)　model　for　predicting　fatigue　lifetime　of　sintered　nano　silver　layer　is　developed.　The　analysis　includes　identifying　the　key　factors　influencing　interconnect　reliability,　and　determining　the　main　factors　affecting　the　thermal　fatigue　lifetime　of　SiC　modules.　The　results　show　that　increasing　the　thickness　of　the　sintered　nano-silver　interconnecting　layer　can　enhance　the　lifetime　of　the　interconnect　structure,　while　elastic　modulus　and　chip　thickness　negatively　impact　the　lifetime　of　the　sintered　silver　layer.　The　prediction　accuracy　and　stability　of　the　proposed　data-driven　SVR　model　are　discussed　and　studied.