33results about How to "In line with wishes" patented technology

The invention is applicable to the technical field of computers, and provides an intelligent response method and device, computer equipment and a storage medium. The intelligent response method comprises the following steps: obtaining dialogue information input by a user; performing intention recognition and keyword matching on the dialogue information; processing the dialogue information by adopting a plurality of response strategy models according to the intention and keywords of the dialogue information to generate a plurality of pieces of response information; analyzing the plurality of pieces of response information, and determining the weight of each piece of response information; and determining the response information with the highest weight as the optimal response information andoutputting the optimal response information. According to the intelligent response method provided by the invention, the intention and the keyword of the dialogue information are recognized firstly,so that the generated multiple pieces of response information are more accurate, the response information is further screened, the optimal response information is screened out and output, and the finally determined response information is more accurate and good in user experience.

The embodiment of the invention provides a method and a device for generating a search prompt. The method for generating the search prompt comprises the following steps: acquiring a search keyword ofa user; generating a candidate prompt set of search keywords according to the search keywords; generating a data feature of each candidate prompt by using a search keyword and each candidate prompt inthe candidate prompt set; inputting the data feature into a pre-trained score prediction model to obtain the score of each candidate prompt; and determining a target prompt for a search keyword fromthe set of candidate prompts according to the score. The score of each candidate prompt is extracted by the score prediction model, No manual weight setting is avoided, the problems of unreasonable weight setting and inaccurate scores of candidate prompts are solved, and the scores of candidate prompt s can be objectively calculated according to the historical behavior of the user, so that the prompts are more in line with the wishes of the user, and the user can select the desired prompts from the prompts, thereby improving the efficiency of information searching.

The invention relates to a method and device for controlling a terminal. The method comprises the steps that whether the tilting state parameter of the terminal meets a preset tilting condition or not is detected; when the tilting state parameter of the terminal meets the tilting condition, whether the terminal rotates around a shaft or not is recognized; when the fact that the terminal rotates around the shaft is recognized, the terminal is triggered to carry out preset operation. Due to the fact that when rotation of the terminal is recognized to be user enabling, the terminal is triggered to carry out preset operation, so that misoperation to the terminal due to the fact that the acceleration variation of the terminal is only judged is reduced, and the accuracy of controlling the terminal and operation is improved.

Embodiments of the present invention disclose a video recommendation method and apparatus, which are applied to a server. The method comprises: acquiring a text string input by a user, and identifying at least one word segment included in the text string; based on text information of each identified word segment, determining a text coefficient corresponding to each word segment; separately determining whether each word segment is the same as a historical search word in a preset weighted word library, and if a determination result is yes, weighting the text coefficient corresponding to the word segment by using a weighting coefficient corresponding to the historical search word that is the same as the word segment, so as to obtain a recommendation coefficient corresponding to the word segment; if the determination result is no, weighting the text coefficient corresponding to the word segment by using a pre-determined weighting coefficient, so as to obtain a recommendation coefficient corresponding to the word segment; normalizing the recommendation coefficient of each word segment; and based on a standard recommendation coefficient that is obtained through normalization and corresponds to each word segment, recommending a video to the user in a preset manner. According to the method and apparatus disclosed by the embodiments of the present invention, video recommendation is more accurate, and user experience is better.

The embodiment of the invention discloses a voice processing method and a voice processing device. A specific embodiment of the method comprises the following steps: carrying out voice recognition onan obtained user voice to obtain a text of the user voice; in response to determining that the text is a reference answer indication statement comprising question and answer pairs, performing generalization processing on the text to obtain a generalization processing result; and correspondingly storing the questions and answers in the question and answer pairs, and storing a generalization processing result. The user in the embodiment of the invention can conveniently and quickly set the reference answer through the voice.

The invention discloses a content management method and device, equipment and a computer readable storage medium. The method comprises the steps of when any content is moved to a target storage area manually by a user, adding specific fields in attribute information corresponding to the content; when the target storage area needs to achieve content moving-back, judging whether or not the specificfields exist in the attribute information corresponding to a content which needs to be moved back, if yes, refusing to move back the content which needs to be moved back, and if no, moving back the content which needs to be moved back. According to the technical scheme, when the target storage area needs to achieve content moving-back, moving-back of the content corresponding to which the attribute information has the added specific fields is refused, moving-back of the content corresponding to which the attribute information does not have the added specific fields is achieved only, accordingly not only can metadata cluster moving-back be achieved, but also moving-back of the content which needs to be stored in the target storage area by a user can be avoided, and thus the will of the usercan be met.

The invention discloses a printing and dyeing process of pure cotton fabric, which is realized through two steps of fabric pretreatment and dyeing. The fabric pretreatment of the present invention combines desizing, scouring and bleaching into one process, reduces the process of the entire printing and dyeing process, improves the cycle of the entire process, improves production efficiency and reduces production costs. Adopting the dyeing method of the invention can make the dyeing more thorough, and the achieved effect is more in line with the wishes of merchants and consumers.

The invention discloses an auxiliary food intake method, system and device and a computer storage medium. The auxiliary food intake method comprises the steps that a stimulation signal corresponding to each piece of food information is displayed; brain waves produced by a user under the stimulation of the stimulation signal corresponding to the food information are acquired; target food information is obtained according to the acquired brain waves, and analysis results are sent to a transmission device so as to make the transmission device select foods selected by the user according to the analysis results; the mouth part including the lips of the user is positioned in real time, and the position information between the positioned mouth part and the transmission device is calculated and issent to the transmission device so as to make the transmission device deliver the selected foods to the mouth part of the user according to the position information. By adopting the auxiliary food intake method, the foods selected by the user more meet the demands of the user, and the technical problem how to help a patient with clumsy hands to automatically eat foods is solved to some degree.

The invention discloses a method for treating avermectin fermentation wastewater and producing a biological feed, relating to the technical field of treatment for biological fermentation and antibiotic fermentation wastewater. The method comprises the following steps: preparing a seed culture medium from glucose, yeast extract, urea, monopotassium phosphate and water, and introducing a geotrichum candidum spore suspension to culture geotrichum candidum seed liquid; fermenting the avermectin, and adding the urea and monopotassium phosphate into the obtained avermectin fermentation wastewater to obtain a geotrichum candidum fermentation culture medium; introducing the geotrichum candidum seed liquid into the geotrichum candidum fermentation culture medium for fermentation culture of geotrichum candidum; and after the fermentation culture of geotrichum candidum, performing filter pressing with a plate, and drying the obtained filter cake to obtain the geotrichum candidum biological feed. Compared with the prior art, the method has the advantages that the wastewater is treated while the biological feed is produced, the water is saved, the technology is simple and practical, the construction cost and operation cost are low, and the method is environmentally friendly and conforms to the industrial policy.

The invention discloses a single plant forming splicing cultivation method for the tree. The method comprises the following steps of S1, selecting trees; S2, cutting off cuticular layers at the same height positions of the two trees, spraying a copper sulfate solution at positions where the cuticular layers are cut off, clasping the two trees together, and then tying the two trees together with straps; S3, smearing a protective agent on the edges of the positions where the cuticular layers are cut off of the two trees, and applying lime to the upper and lower sides of the positions 10 cm awayfrom the positions where the cuticular layers are cut off of the two trees; S4, spraying a nutrient solution and a hormone mixture to the places where the cuticular layers are cut off of the two trees, and applying nutrient soil to the roots of the two trees simultaneously. According to the method, by splicing the two trees, different trees can be spliced into one plant, the spliced trees have good styling effect and richer variety, the trees can be grown and formed according to the design of people, and the splicing of the trees is more in line with wishes of the people.

The invention relates to the field of music generation, in particular to a music generation method based on variable neighborhood search and a masking effect. The problem that as for an existing method, music generation cannot be conducted according to music clips obtained through the sudden inspiration or extemporaneous play of a composer, and the problem that generated music may have cluttered notes are solved. Time domain training data are subjected to melody extraction based on the masking effect, mapped to a note domain and visualized; a to-be-trained music score is input into a model toobtain a parameter matrix, and the parameter matrix can be mixed with existing training model parameters; at the generation stage, based on a variable neighborhood search algorithm, the music score isiterated repeatedly before meeting the threshold, and grading is conducted; and finally, subsequent sections or accompaniments and drumbeats can be selected to be generated. The music score can be manually corrected and annotated nearly every step, and the threshold of composing can be lowered while repeatable work is reduced; and the music generation method is mainly used for music generation.

The invention discloses an anti-mating device for dogs, and relates to the field of dog supplies. The problem that at present, dogs or cats conduct random mating, and trouble is caused is solved. Theanti-mating device for the dogs is provided with a shading net, a strap, abdominal belts, a waistband and a fixing band, the shading net is a net cover shaped like an oval, the mesh diameter is smaller than 0.5 cm, the abdominal belts are fixed to the two corners of the lower end of the shading net respectively, and the two corners of the upper end of the shading net are fixed to two traction ropes on the strap respectively; the fixing band is movably connected between a dog ring and the waistband. The device is simple in structure, easy to manufacture, convenient to use, safe and reliable.