Method, device and computer program product for updating user interface
Embodiments of the present disclosure provide a method, device and computer program product for updating a user interface. According to example implementations of the present disclosure, an element sequence including a plurality of elements in the user interface is obtained, each element in the element sequence being associated with each of a plurality of actions being performed by a user in the user interface, the plurality of elements in the element sequence being sorted in an order of the plurality of actions being performed by the user; a natural language processing model is trained using the element sequence, the natural language processing model being used for modeling and feature-learning of a natural language; and the user interface is enabled to be updated based on the trained natural language processing model. Therefore, software developers can have deeper insight into users' needs and develop a more user-friendly user interface.
1. A method of updating a user interface, comprising:
generating an action dictionary represented by a matrix based in part on an element identifier assigned to each of a plurality of elements in the user interface, the generated action dictionary comprising a set of a pair of the element identifier and a corresponding element of the plurality of elements in the user interface, wherein the element identifier is assigned according to a hierarchical structure of the user interface, wherein the user interface includes a plurality of elements, each of the plurality of elements include a plurality of sub-elements;
obtaining an element sequence including the plurality of elements in the user interface, each element in the element sequence being associated with each of a plurality of actions being performed by a user in the user interface, the plurality of elements in the element sequence being sorted in an order of the plurality of actions being performed by the user;
training a natural language processing model using the element sequence, the natural language processing model being used for modeling and feature-learning of a natural language, wherein the training is based on a Huffman tree and includes iteratively updating an input feature vector and a plurality of vectors of an internal node in the Huffman tree for the internal node;
determining a plurality of feature vectors from the trained natural language processing model, each of the plurality of feature vectors corresponding to each of the plurality of elements in the element sequence and indicating a position of each element in a semantic space; and
determining, based on the plurality of feature vectors, relations between the positions of the plurality of elements in the user interface and the plurality of elements; and
enabling the user interface to be updated based on the trained natural language processing model, wherein the plurality of elements in the updated user interface are different than the plurality of elements in the user interface in terms of position or structure.
2. The method of claim 1 , wherein obtaining the element sequence comprises:
obtaining the plurality of actions being performed by the user in the user interface;
determining the plurality of elements associated with the plurality of actions;
determining an identifier pre-assigned to each of the plurality of elements; and
sorting the determined identifiers in the order of the plurality of actions being performed by the user.
3. The method of claim 1 , wherein an interval between an action associated with a last element in the element sequence and an action associated with a first element in a next element sequence exceeds a predetermined threshold time.
4. The method of claim 1 , wherein obtaining the element sequence comprises:
obtaining a user identifier associated with the plurality of actions, the user identifier being preset for the user performing the plurality of actions; and
obtaining, based on the user identifier, the element sequence for the user.
5. The method of claim 1 , wherein updating the user interface comprises:
obtaining a current element associated with a current action being performed by the user in the user interface;
determining, based on the current element and the trained natural language processing model, a predicting element associated with an action to be potentially performed by the user after the current action; and
indicating the predicting element in the user interface.
6. A device for updating a user interface, comprising:
at least one processing unit; and
at least one memory coupled to the at least one processing unit and storing instructions to be executed by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the device to perform acts including:
generating an action dictionary represented by a matrix based in part on an element identifier assigned to each of a plurality of elements in the user interface, the generated action dictionary comprising a set of a pair of the element identifier and a corresponding element of the plurality of elements in the user interface, wherein the element identifier is assigned according to a hierarchical structure of the user interface, wherein the plurality of elements in the updated user interface are different than the plurality of elements in the user interface in terms of position or structure;
obtaining an element sequence including the plurality of elements in the user interface, each element in the element sequence being associated with each of a plurality of actions being performed by a user in the user interface, the plurality of elements in the element sequence being sorted in an order of the plurality of actions being performed by the user;
training a natural language processing model using the element sequence, the natural language processing model being used for modeling and feature-learning of a natural language, wherein the training is based on a Huffman tree and includes iteratively updating an input feature vector and a plurality of vectors of an internal node in the Huffman tree for the internal node;
determining a plurality of feature vectors from the trained natural language processing model, each of the plurality of feature vectors corresponding to each of the plurality of elements in the element sequence and indicating a position of each element in a semantic space; and
determining, based on the plurality of feature vectors, relations between the positions of the plurality of elements in the user interface and the plurality of elements; and
enabling the user interface to be updated based on the trained natural language processing model, wherein the plurality of elements in the updated user interface are different than the plurality of elements in the user interface in terms of position or structure.
7. The device of claim 6 , wherein obtaining the element sequence comprises:
obtaining the plurality of actions being performed by the user in the user interface;
determining the plurality of elements associated with the plurality of actions;
determining an identifier pre-assigned to each of the plurality of elements; and
sorting the determined identifiers in the order of the plurality of actions being performed by the user.
8. The device of claim 6 , wherein an interval between an action associated with a last element in the element sequence and an action associated with a first element in a next element sequence exceeds a predetermined time threshold.
9. The device of claim 6 , wherein obtaining the element sequence comprises:
obtaining a user identifier associated with the plurality of actions, the user identifier being preset for the user performing the plurality of actions; and
obtaining, based on the user identifier, the element sequence for the user.
10. The device of claim 6 , wherein updating the user interface comprises:
obtaining a current element associated with a current action being performed by the user in the user interface;
determining, based on the current element and the trained natural language processing model, a predicting element associated with an action to be potentially performed by the user after the current action; and
indicating the predicting element in the user interface.
11. A computer program product, tangibly stored on a non-transient computer readable medium and comprising machine executable instructions which, when executed, cause a machine to perform operations, the operations comprising:
generating an action dictionary represented by a matrix based in part on an element identifier assigned to each of a plurality of elements in a user interface, the generated action dictionary comprising a set of a pair of the element identifier and a corresponding element of the plurality of elements in the user interface, wherein the element identifier is assigned according to a hierarchical structure of the user interface, wherein the plurality of elements in the updated user interface are different than the plurality of elements in the user interface in terms of position or structure;
obtaining an element sequence including the plurality of elements in the user interface, each element in the element sequence being associated with each of a plurality of actions being performed by a user in the user interface, the plurality of elements in the element sequence being sorted in an order of the plurality of actions being performed by the user;
training a natural language processing model using the element sequence, the natural language processing model being used for modeling and feature-learning of a natural language, wherein the training is based on a Huffman tree and includes iteratively updating an input feature vector and a plurality of vectors of an internal node in the Huffman tree for the internal node;
determining a plurality of feature vectors from the trained natural language processing model, each of the plurality of feature vectors corresponding to each of the plurality of elements in the element sequence and indicating a position of each element in a semantic space; and
determining, based on the plurality of feature vectors, relations between the positions of the plurality of elements in the user interface and the plurality of elements; and
enabling the user interface to be updated based on the trained natural language processing model, wherein the plurality of elements in the updated user interface are different than the plurality of elements in the user interface in terms of position or structure.
12. The computer program product of claim 11 , wherein obtaining the element sequence comprises:
obtaining the plurality of actions being performed by the user in the user interface;
determining the plurality of elements associated with the plurality of actions;
determining an identifier pre-assigned to each of the plurality of elements; and
sorting the determined identifiers in the order of the plurality of actions being performed by the user.
13. The computer program product of claim 11 , wherein an interval between an action associated with a last element in the element sequence and an action associated with a first element in a next element sequence exceeds a predetermined threshold time.
14. The computer program product of claim 11 , wherein obtaining the element sequence comprises:
obtaining a user identifier associated with the plurality of actions, the user identifier being preset for the user performing the plurality of actions; and
obtaining, based on the user identifier, the element sequence for the user.
15. The computer program product of claim 11 , wherein updating the user interface comprises:
obtaining a current element associated with a current action being performed by the user in the user interface;
determining, based on the current element and the trained natural language processing model, a predicting element associated with an action to be potentially performed by the user after the current action; and
indicating the predicting element in the user interface.