Abstract: In solving the problems of small-size samples, non-linear and high-dimensional pattern recognitions, SVM exhibits many unique advantages. In this paper, for the first time, SVM method is introduced for the selection of leading industries. Based on development trend benchmarks, industrial relational degree benchmarks, innovation benchmarks, regional comparative advantage and sustainable development benchmarks, a regional leading industry index system that incorporates 15 indicators is constructed, applying SVM method to quantitatively study 43 double-digit industries of Chongqing to find its leading industries. These are transport equipment manufacturing industry, non-ferrous metal smelting and rolling processing industry, oil and gas exploration industry, transportation, warehousing and postal industries, construction, communications equipment, computers and other electronic equipment manufacturing, wholesale and retail trade, accommodation and catering industry. The research results of Chongqing are in basic congruity with its current situation and future development plans, demonstrating that SVM method is scientific, rational and forward-looking.