The emergence and development of shared E-bicycles (SEB) has resulted in complex connection and substitution relationships with public transportation (PT) in urban areas, impacting the social benefits of integrated transportation systems and the return on investment of PT subsidies. Accurately identifying how SEB complement and/or compete with PT is still a crucial challenge. Therefore, this research proposes a new discrimination method based on multi-source data fusion to address this issue. The method’s discrimination accuracy is evaluated based on both telephone interviews data and transaction data from SEB users in Shenyang City, China. The results indicate that SEB are primarily used for single-mode trips (75%). When SEB are used to combine with other modes, they mainly serve as connecting modes to rail transit (22.1%). On the other hand, shared bicycles (SB) and buses are the two modes most replaced by SEB, followed by private cars and walking. 63% of SEB rides can be accurately classified by the proposed method in terms of their connection and substitution relationship with PT. Incorporating anonymized socioeconomic attribute information from big data holds promise for further enhancing the method’s performance. This research provides key insights for assessing the societal benefits of SEBs on PT, offering valuable theoretical and methodological support for transportation planning.