Systems and methods for the analysis of user experience testing with AI acceleration
Systems and methods for AI assisted analysis of a user experience study are provided. A study objective (a goal of the study) and data relating to all possible navigation routes within a digital interface are received. Simulated clickstreams for navigating from any state of the digital interface to the study objective are generated. This simulated clickstream data is then used to train one or more machine learning models to determine a most efficient path to achieve the study objective from any state of the digital interface. Subsequently, study results from many different participants is received. Key events are then identified within the study results. Additionally, the likelihood of failure for each of the plurality of study results is predicted using the machine learning model, and information density of the plurality of study results is determined.
1. A method for artificial intelligence artificial intelligence (AI) assisted analysis of a user experience study comprising:
receiving a study objective and data relating to all possible navigation routes within a, digital interface;
receiving at least one recording of a study executed by at least one individual for the digital interface;
training a machine learning model using the recording to determine how a user behaves when presented with a given state within the digital interface;
receiving at least one version of a prototype interface related to the digital interface;
predicting likelihood of failure for each of the at least one version of a prototype interface using the machine learning model; and
providing improvement suggestions for the at least one prototype interface responsive to the predicted likelihood of failure.
2. The method of claim 1 , further comprising generating a plurality of simulated clickstreams includes a series of search methods to explore state space within the at least one prototype interface, wherein the generating the plurality of simulated clickstreams is performed as an asynchronous batch dataset, and is iteratively performed.
3. The method of claim 1 , wherein the machine learning model is trained by reinforcement learning algorithms.
4. The method of claim 1 , wherein the machine learning model includes a distance model from each state of the web page to the study objective, wherein the distance model is a number of actions required to reach the study objective from a particular state.
5. The method of claim 4 , wherein the distance model is a number of actions weighted by empirically measured time of each action, or frequency probability of each action, required to reach the study objective from a particular state.
6. The method of claim 1 , further comprising identifying key events by video recognition of a video recording, audio analysis of an audio recording and clickstream data in combination with a known goal, to identify emotions of interest and keyword analysis.
7. The method of claim 1 , further comprising:
receiving failure predictions and information density for a plurality of studies;
ranking the plurality of studies by the information density;
clustering the plurality of studies by failure type and location; and
rendering analysis results.