With　the　rapid　development　of　network　technology,　network　security　problems　are　gradually　increasing,　and　the　network　attack　situation　is　very　severe.　In　a　complex　attack　scenario,　timely　detection　of　potential　attack　behaviors　and　timely　identification　and　pre-judgment　of　attack　intentions　are　important　components　of　security　risks.　However,　the　attack　behavior　in　the　network　presents　complexity,　multi-step　and　uncertainty,　which　brings　new　technical　challenges　to　attack　intent　analysis.　Aiming　at　the　problem　that　the　attack　intention　of　multi-step　complex　attack　is　difficult　to　identify,　this　paper　proposes　an　attack　intention　analysis　method　based　on　attack　path　graph.　Firstly,　aiming　at　the　multi-step　complex　attack　behavior　analysis　problem,　the　key　asset　assessment　technology　is　used　to　find　out　the　key　assets　in　the　network　system,　and　the　hypothetical　attack　intention　is　generated　according　to　the　security　protection　requirements　of　the　network　system.　Then,　it　is　difficult　to　manually　construct　the　attack　path　map　in　the　large-scale　network,　and　the　automatic　generation　of　the　attack　path　map　is　realized.　Finally,　a　method　of　network　attack　intent　identification　is　proposed　and　a　calculation　method　of　attack　intent　probability　is　designed,　which　improves　the　efficiency　and　accuracy　of　attack　intent　recognition.　©　2019　Association　for　Computing　Machinery.