The development of the Internet and the Internet of Things has led to a sharp increase in the amount of data. The great advantages of big data have promoted the research and usage of a series of related technologies in various fields. In the field of government, data from various departments is being aggregated to acquire more value which can help to improve the efficiency and quality of public service. Cloud computing, as an infrastructure of big data, is also applied in government big data. However, the sensitivity of government data determines that the data stored in the cloud must be well protected. Meanwhile, fine-grained data sharing is also important for the public services of government. Recently, several searchable attribute-based encryption schemes have been proposed to achieve fine-grained data access control and search on ciphertext simultaneously. Unfortunately, each of them has some imperfections in efficiency or access policy. In this paper, we propose a fine-grained authorized keyword secure search scheme by leveraging the attribute-based encryption primitive, whose access policy supports AND, OR, and threshold gates. We give the concrete construction, rigorous verification of correctness, detailed security analysis, and prove that our solution is efficiency through several experiments.