This paper addresses the problem of joint transmission and computation resource allocation in a multi-user probabilistic semantic communication (PSC) network. Users compress their large-sized data using semantic information extraction techniques and transmit the compressed data to a multi-antenna base station (BS). The model represents large-sized data through extensive knowledge graphs and compresses it using shared probability graphs between users and the BS. The resource allocation problem is formulated as an optimization problem to maximize the sum of equivalent rates for all users, considering total power and semantic resource constraints. To solve this non-convex and non-smooth optimization problem, a three-stage algorithm is proposed: stage 1 optimizes the receive beamforming matrix using the minimum mean square error (MMSE) strategy, stage 2 performs a rough search for the semantic compression ratio using an alternating optimization (AO) method, and stage 3 refines the semantic compression ratio using gradient ascent. Numerical results validate the effectiveness of the proposed scheme, demonstrating its ability to achieve a balanced equilibrium between transmission and computation.This paper addresses the problem of joint transmission and computation resource allocation in a multi-user probabilistic semantic communication (PSC) network. Users compress their large-sized data using semantic information extraction techniques and transmit the compressed data to a multi-antenna base station (BS). The model represents large-sized data through extensive knowledge graphs and compresses it using shared probability graphs between users and the BS. The resource allocation problem is formulated as an optimization problem to maximize the sum of equivalent rates for all users, considering total power and semantic resource constraints. To solve this non-convex and non-smooth optimization problem, a three-stage algorithm is proposed: stage 1 optimizes the receive beamforming matrix using the minimum mean square error (MMSE) strategy, stage 2 performs a rough search for the semantic compression ratio using an alternating optimization (AO) method, and stage 3 refines the semantic compression ratio using gradient ascent. Numerical results validate the effectiveness of the proposed scheme, demonstrating its ability to achieve a balanced equilibrium between transmission and computation.