As AWOL pointed out, some special equipment is needed to test the sensor. Arduino is not able to provide the capability. From further reading, I think I know what this is about: oscillating the sensor at around 1MH, which is its resonant frequency, and detecting the shift of this resonant frequency when liquid concentration changes. The higher the liquid density (due to concentration change), the more mass the oscillator has to interact in order to oscillate, thus the lower the resonant frequency.
I've only done low frequency resonant circuits so I'm projecting my experience beyond my comfort zone:
You need a source that generates the frequency at 1MHz and the voltage required (400V?). This source needs to be able to sweep frequency. You also need a means to monitor the current of the circuit to detect maximal current flow (near resonant frequency). You then feed the sweeping frequency to the sensor and immerse it in liquid for testing. The source should have a digital output in sync of the frequency sweep to trigger an oscilloscope. Then you convert the circuit's current into voltage and plot it on the scope. The sweep rate and scope time axis tells you the resonant frequency. The graph is a flat line and a sharp peak. As you vary liquid concentration, the sharp peak moves as a response.
Does the above sound familiar to the OP or anyone else here that have ACTUALLY done resonant circuits around 1MHz?