Abstract: A method for training a contrastive learning model, acquiring an optimization item, and pushing information includes: obtaining sample data including first category information and query information; predicting a correlation between the first category information and the query information using the contrastive learning model; establishing a loss function based on the prediction results, wherein the loss function is configured to characterize accuracy of the prediction results; optimizing the loss function based on a semantic relationship between the first category information and second category information, wherein the first category information and the second category information correspond to different semantic information in a category tree, and the semantic relationship is related to relative positions of the first category information and the second category information within the category tree; and training the contrastive learning model based on the optimized loss function.

Abstract: The present invention discloses a ceramic ball automatic sorting system and method. The system automatically sucks a ceramic ball on a ceramic ball feeding track for image acquisition, identifies whether the ceramic ball is defective according to the acquired image information, and determines a ball storage device into which the ceramic ball is placed, and the whole process does not require manual participation, which achieves the automation of ceramic ball defect identification and sorting and improves the ceramic ball defect identification accuracy and sorting efficiency. According to the method, images of the ball surface shot at each angle are automatically spliced by using an automatic image splicing technology to achieve full coverage.

Abstract: The present invention discloses a ceramic ball automatic sorting system and method. The system automatically sucks a ceramic ball on a ceramic ball feeding track for image acquisition, identifies whether the ceramic ball is defective according to the acquired image information, and determines a ball storage device into which the ceramic ball is placed, and the whole process does not require manual participation, which achieves the automation of ceramic ball defect identification and sorting and improves the ceramic ball defect identification accuracy and sorting efficiency. According to the method, images of the ball surface shot at each angle are automatically spliced by using an automatic image splicing technology to achieve full coverage.

Abstract: A composition and a method for powder injection moulding are provided. The composition includes a powder and a binder. The binder includes an alkane-based primary main binder and a multifunctional secondary main binder. The alkane-based primary main binder composes 20 mass % or more of the binder. The multifunctional secondary main binder is at least partially soluble in the alkane-based primary binder, is soluble in water, is thermally evaporable, possesses a dispersant functionality towards the powder, and increases a feedstock viscosity of the composition at a near zero feedstock shear rate.

Abstract: A method for providing an advertisement is disclosed. The method, to be implemented on a machine having at least one processor, storage, and a communication platform connected to a network may include receiving a request for an advertisement to be inserted into a content page to be presented to a user. The method further includes selecting an advertisement relevant to the content page and the user based on a content-advertisement index and a user-advertisement index, constructed based on user browsing context, and transmitting information indicative of the selected advertisement as a response to the request. Further disclosed is a method for generating multi-dimensional index.

Abstract: The present disclosure provides a system and method for suggesting a bidding keyword to an advertiser. The system includes a non-transitory processor-readable storage medium comprising a set of instructions for suggesting a bidding keyword to an advertiser; and a processor in communication with the storage medium. The processor is configured to execute the set of instruction to receive an advertisement creative from an advertiser; determine, based on the advertisement creative without using an externally input seed keyword, a recommended bidding keyword associated with the advertisement creative; and return the recommended keyword for online advertisement bidding.

Abstract: The present invention belongs to the field of technology known as powder injection moulding. It includes producing a part by powder injection moulding at low to high injection pressure and it discloses a novel binder system for optimized feedstock design. The primary subject matter of the invention resides in introducing a secondary main binder into the feedstock binder composition, meeting, simultaneously, the five (5) following criteria: partially soluble in the primary binder, soluble in water, thermally evaporable, possessing a dispersant functionality and increasing the viscosity at near zero feedstock shear rate. In addition to this binder, the feedstock may contain a small fraction of another dispersant and a back-bone polymer when needed, depending on the type of powder material used for the feedstock. Based on the main modes of binder removal (thermal or water extraction) this novel binder system is denominated as Te-Wex™ system.

Abstract: Systems and methods are provided for search results targeting. The system includes a first database including advertiser bidding information and a second database including websites statistics generated by a search engine. The system includes an input from an advertiser and a plurality of advertiser keywords obtained from the first database system based on the input from the advertiser. The system includes one or more modules configured to: rank the plurality of advertiser keywords; obtain a plurality of website identifiers for each top ranked advertiser keywords from a second database; rank the obtained website identifiers based on history statistics of the obtained website identifiers; and select the top ranked website identifiers as retargeting candidates for the advertiser.