Alzheimer's disease (AD) is usually accompanied by different degrees of behavioral and psychological symptoms of dementia (BPSD). Transcranial magnetic stimulation (TMS) has been applied for the treatment of AD as a painless and noninvasive therapy. However, the efficacy of repetitive TMS (rTMS) with different frequencies in AD with BPSD remains unknown. A total of 32 AD patients with psychobehavioral symptoms were selected as the study subjects. Among them, 16 patients were included in the high-frequency TMS group with an average disease duration of 6.22 ± 2.55years. The low-frequency TMS group was gender and age matched with a disease course of 7.02 ± 3.33-year average duration. The high-frequency TMS group received TMS treatment twice per day for 4weeks under 80% MT stimulation intensity, 10-Hz frequency for 0.5h each time, and the low-frequency TMS group received TMS treatment of 2-Hz frequency for 0.5h each time. Neuropsychological status was assessed by the Behavioral Pathology in Alzheimer's Disease Rating Scale (BEHAVE-AD) score. The behavioral ability was assessed by the Abilities of Daily Living (ADL) scale; cognitive function was evaluated by Mini-Mental State Examination (MMSE). The levels of β amyloid 40 and 42 (Aβ40 and Aβ42) in plasma were detected using a double-antibody sandwich enzyme-linked immunosorbent assay. All patients underwent brain magnetic resonance imaging (MRI) before and after the experiment. After 2weeks of treatment, the BEHAVE-AD and ADL scores of the patients in the high-frequency group were significantly lower than those before the treatment, and they continued to decrease after 4weeks of treatment. The BEHAVE-AD and ADL scores of the low-frequency TMS group were also significantly lower than before treatment. The comparison between groups at different time points showed that the BEHAVE-AD and ADL scores of the patients in the high-frequency group were significantly lower than those of the patients in the low-frequency TMS group. The MMSE of high-frequency TMS-treated patients increased from 14.22 ± 3.55 before treatment to 14.67 ± 2.22 at 2week's treatment and 17.33 ± 3.11 at 4week's treatment (p < 0.01) in contrast to 14.19 ± 3.47, 14.28 ± 3.41, and 14.49 ± 2.79, respectively, found in the low-frequency TMS group. At week 4, the high-frequency TMS-treated group's plasma Aβ40 did not change compared to that before treatment. No effects on plasma Aβ42 were observed between the high- vs. low-frequency TMS groups. The incidence of adverse reactions during treatment was comparable between groups. These results indicate that high-frequency TMS has the advantages of fast results, good efficacy, and high safety for the treatment of psychobehavioral abnormalities in AD patients. In addition, our study suggests that high-frequency TMS intervention can further improve the cognitive function of AD patients.