Abstract: A polyurethane is obtained by continuously reacting a polyurethane prepolymer in an asphalt or asphalt mixture material system under high temperature. The prepolymer is prepared by: adding a polymerization inhibitor, catalyst and isocyanate component in a reaction container, adding a polyol to the reactor while stirring at room temperature under nitrogen, increasing the temperature to 50-80° C., and maintaining the temperature to react for 0.5-6 hours. A polyurethane modified asphalt, a mixture material containing the polyurethane and a polyurethane modified asphalt pavement structure can be prepared. The synthesis condition of the polyurethane is mild, and a secondary reaction with the air and an active hydrogen component in the asphalt can occur during maintenance and formation to further increase the strength of a mixture material. The standard Marshall stability at 60° C. meets the petroleum asphalt requirement, such that a service life of the asphalt pavement can be increased.

Abstract: A thermoset modified additive for asphalt mixture includes the following raw materials: resin, curing agent, promoter, flow agent, filler and additive, wherein the weight part proportions of the raw materials are: resin 35-70, curing agent 1-35, promoter 0-2.5, flow agent 0-6, filler 20-40 and additive 0-2. The asphalt mixture includes thermoset modified additive for asphalt mixture, matrix asphalt or modified asphalt and mineral aggregate, wherein the weight part proportions of the raw materials are: thermoset modified additive for asphalt mixture 1-5, matrix asphalt 2-8 and mineral aggregate 87-97. The thermoset modified additive for asphalt mixture is prepared by dry process modification technology; the prepared modified additive is powdery, has small particle size, and is easy to disperse in asphalt mixture; and the asphalt mixture prepared from the modified additive has simple operation process and low cost, and can be subjected to continuous production.

Abstract: This present invention discloses a waterborne polymer modified emulsified asphalt mixture and the preparation method thereof, and particularly relates to a waterborne polyurethane emulsified asphalt concrete, a waterborne acrylic resin emulsified asphalt concrete, and a waterborne epoxy resin emulsified asphalt micro-surfacing mixture, and preparation methods thereof. A mixture containing a waterborne polymer modified emulsified asphalt forms a high-performance composite system having a spatial network structure, and has good performance and simple preparation process.

Abstract: The disclosure provides bio-asphalt, bio-asphalt compositions and a method for preparing them. It belongs to the technical field of asphalt material. The method of the disclosure uses bio-oil as the raw material, adopts a combination of extraction process and oxidation process to prepare bio-asphalt, and modifies the product obtained from bio-oil by extraction process or bio-asphalt with ordinary petroleum asphalt and/or aromatic hydrocarbon oil and a macromolecular material to obtain a bio-asphalt composition.

Abstract: A thermoset modified additive for asphalt mixture includes the following raw materials: resin, curing agent, promoter, flow agent, filler and additive, wherein the weight part proportions of the raw materials are: resin 35-70, curing agent 1-35, promoter 0-2.5, flow agent 0-6, filler 20-40 and additive 0-2. The asphalt mixture includes thermoset modified additive for asphalt mixture, matrix asphalt or modified asphalt and mineral aggregate, wherein the weight part proportions of the raw materials are: thermoset modified additive for asphalt mixture 1-5, matrix asphalt 2-8 and mineral aggregate 87-97. The thermoset modified additive for asphalt mixture is prepared by dry process modification technology; the prepared modified additive is powdery, has small particle size, and is easy to disperse in asphalt mixture; and the asphalt mixture prepared from the modified additive has simple operation process and low cost, and can be subjected to continuous production.

Abstract: The disclosure provides bio-asphalt, bio-asphalt compositions and a method for preparing them. It belongs to the technical field of asphalt material. The method of the disclosure uses bio-oil as the raw material, adopts a combination of extraction process and oxidation process to prepare bio-asphalt, and modifies the product obtained from bio-oil by extraction process or bio-asphalt with ordinary petroleum asphalt and/or aromatic hydrocarbon oil and a macromolecular material to obtain a bio-asphalt composition.

Abstract: This present invention discloses a waterborne polymer modified emulsified asphalt mixture and the preparation method thereof, and particularly relates to a waterborne polyurethane emulsified asphalt concrete, a waterborne acrylic resin emulsified asphalt concrete, and a waterborne epoxy resin emulsified asphalt micro-surfacing mixture, and preparation methods thereof. A mixture containing a waterborne polymer modified emulsified asphalt forms a high-performance composite system having a spatial network structure, and has good performance and simple preparation process.

Abstract: A machine-learning method for estimating probability of a click event in online advertising systems by computing and comparing an aggregated predictive model (a global model) and one or more data-wise sliced predictive models (local models). The method comprises receiving training data having a plurality of features stored in a feature set and constructing a global predictive model that estimates the probability of a click event for the processed feature set. Then, partitioning the global predictive model into one or more data-wise sliced training sets for training a local model from each of the data-wise slices, and then determining whether a particular local model estimates probability of click event for the feature set better than the global model. A given feature set may be collected from historical data, and may comprise a feature vector for a plurality of query-advertisement pairs and a corresponding indicator that represents a click on the advertisement.

Abstract: A machine-learning method for estimating probability of a click event in online advertising systems by computing and comparing an aggregated predictive model (a global model) and one or more data-wise sliced predictive models (local models). The method comprises receiving training data having a plurality of features stored in a feature set and constructing a global predictive model that estimates the probability of a click event for the processed feature set. Then, partitioning the global predictive model into one or more data-wise sliced training sets for training a local model from each of the data-wise slices, and then determining whether a particular local model estimates probability of click event for the feature set better than the global model. A given feature set may be collected from historical data, and may comprise a feature vector for a plurality of query-advertisement pairs and a corresponding indicator that represents a click on the advertisement.