摘要：【中文】本发明提供一种社交关系的挖掘方法及装置、计算机设备与可读介质。其方法包括：将采集到的多个应用的用户行为数据基于用户进行打通，获取各个用户的行为数据；基于各用户的行为数据，挖掘数对双向关系用户对。与现有的社交关系挖掘方案相比，本发明的技术方案，不局限于仅对一个应用内的社交关系进行挖掘，可以适用于对多个应用中的社交关系进行挖掘，具有非常强的实用性，同时还能够保证挖掘的社交关系的准确性。 【EN】The invention provides a social relationship mining method and device, computer equipment and a readable medium. The method comprises the following steps: making a call on the collected user behavior data of the plurality of applications based on the user to obtain the behavior data of each user; and mining the number pair bidirectional relation user pairs based on the behavior data of each user. Compared with the existing social relationship mining scheme, the technical scheme of the invention is not limited to mining the social relationship in one application, can be suitable for mining the social relationship in multiple applications, has very strong practicability, and can ensure the accuracy of the mined social relationship.

摘要：【中文】本申请提出一种知识图谱生成方法、装置、计算机设备以及存储介质，其中，方法包括：通过对包含多次搜索的搜索会话，获取在搜索会话中用户输入的各搜索词，进而根据各搜索词之间的文本重合情况或各搜索词的句法结构，确定各搜索词之间的语义包含关系，将各搜索词作为知识图谱节点，根据各搜索词之间的语义包含关系，确定各节点之间的父子关系，最终根据父子关系，生成知识图谱。该方法通过用户搜索会话输入的各搜索词，根据各搜索词之间的文本重合情况或各搜索词的句法结构，确定各搜索词之间的语义包含关系，进而生成知识图谱，解决了现有技术中不能准确识别语义包含关系以及通过人工的方式构建知识图谱，导致构建成本高，时间长的技术问题。 【EN】The application provides a knowledge graph generation method, a knowledge graph generation device, computer equipment and a storage medium, wherein the method comprises the following steps: the method comprises the steps of obtaining search words input by a user in a search session through the search session containing multiple searches, determining semantic inclusion relation among the search words according to text superposition conditions among the search words or syntax structures of the search words, using the search words as knowledge graph nodes, determining parent-child relation among the nodes according to the semantic inclusion relation among the search words, and finally generating a knowledge graph according to the parent-child relation. The method determines semantic inclusion relation among the search words according to text superposition conditions among the search words or syntax structures of the search words by searching the search words input by a user, and further generates the knowledge graph, thereby solving the technical problems of high construction cost and long construction time caused by the fact that the semantic inclusion relation cannot be accurately identified and the knowledge graph cannot be constructed in a manual mode in the prior art.

摘要：【中文】本申请公开了一种数据分析方法、装置、设备和计算机存储介质，涉及大数据领域。具体实现方案为：从目标应用的用户行为数据中，获取流失用户的行为数据；从所述流失用户的行为数据中提取用户行为特征，构成各流失用户的特征向量；基于特征向量之间的距离对所述流失用户进行聚类，得到一个以上的用户簇；分别针对各用户簇分析主成分特征，以确定各用户簇的主要流失原因。本申请能够实现低成本、高准确性地分析用户流失原因。 【EN】The application discloses a data analysis method, a data analysis device, data analysis equipment and a computer storage medium, and relates to the field of big data. The specific implementation scheme is as follows: acquiring behavior data of the lost user from the user behavior data of the target application; extracting user behavior characteristics from the behavior data of the lost users to form characteristic vectors of all lost users; clustering the lost users based on the distance between the feature vectors to obtain more than one user cluster; and analyzing the principal component characteristics of each user cluster respectively to determine the main loss reason of each user cluster. The method and the device can analyze the user loss reason with low cost and high accuracy.

摘要：【中文】本申请公开了用户点击模型的训练方法、装置、电子设备及存储介质，涉及人工智能领域。具体实现方案为：从全网用户的日志数据库中采集数条行为数据，各条所述行为数据中包括预设时间周期内一用户对网络中的资源的反馈信息；基于预先创建的头部资源库和所述数条行为数据，生成所述数条行为数据对应的数条资源访问特征；采用所述数条资源访问特征，训练用户点击模型。本申请能够提供一种无损的建模方式，与现有技术的建模方式相比，能够有效地优化用户点击模型的精度，提高用户点击模型的准确性。 【EN】The application discloses a training method and device for a user click model, electronic equipment and a storage medium, and relates to the field of artificial intelligence. The specific implementation scheme is as follows: collecting a plurality of pieces of behavior data from a log database of users in the whole network, wherein each piece of behavior data comprises feedback information of one user to resources in the network within a preset time period; generating a plurality of resource access characteristics corresponding to the plurality of pieces of behavior data based on a pre-created head resource library and the plurality of pieces of behavior data; and training a user click model by adopting the plurality of resource access characteristics. Compared with the modeling mode in the prior art, the method and the device can effectively optimize the precision of the user click model and improve the accuracy of the user click model.

摘要：【中文】本申请公开了一种语音合成的评估方法、装置、电子设备及可读存储介质，涉及语音技术领域。具体实现方案为：采集测试用户收听智能对话系统的合成语音后，对经过合理性验证的至少两个评估维度的各所述评估维度上的各评估指标的打分；其中，一个所述评估维度上包括至少一个所述评估指标；根据预先拟合出的各所述评估维度上的各所述评估指标的权重以及各所述评估维度上的各所述评估指标的打分，计算各所述评估维度的综合分值。本申请的技术方案，能够弥补现有技术的不足，实现从不同维度上对语音合成进行评估，进而能够从多个评估维度上对语音合成进行更加全面和详细地评估。 【EN】The application discloses a method and a device for evaluating voice synthesis, electronic equipment and a readable storage medium, and relates to the technical field of voice. The specific implementation scheme is as follows: after a test user listens to the synthesized voice of the intelligent dialogue system, scoring each evaluation index on each evaluation dimension of at least two evaluation dimensions subjected to rationality verification; wherein one of the evaluation dimensions comprises at least one of the evaluation metrics; and calculating a comprehensive score of each evaluation dimension according to the weight of each evaluation index in each evaluation dimension and the score of each evaluation index in each evaluation dimension which are fitted in advance. According to the technical scheme, the defects of the prior art can be made up, the speech synthesis can be evaluated from different dimensions, and the speech synthesis can be evaluated more comprehensively and in detail from a plurality of evaluation dimensions.

摘要：【中文】本申请公开了一种获取兴趣点(POI)状态信息的方法、装置、设备和计算机存储介质，涉及大数据领域。具体实现方案为：从互联网获取预设时段内包含POI信息的文本；利用预先训练得到的POI状态识别模型对所述文本进行识别，得到所述文本中的二元组，所述二元组包括POI名称和该POI名称对应的POI状态信息；其中所述POI状态识别模型对所述文本对应的词序列分别进行POI名称的标签预测和POI状态的标签预测，将POI名称的标签预测结果和POI状态的标签预测结果进行对齐，得到所述二元组。本申请能够节约人力成本，提高时效性和准确性。 【EN】The application discloses a method, a device, equipment and a computer storage medium for acquiring point of interest (POI) state information, and relates to the field of big data. The specific implementation scheme is as follows: acquiring a text containing POI information in a preset time period from the Internet; recognizing the text by using a POI state recognition model obtained by pre-training to obtain a binary group in the text, wherein the binary group comprises a POI name and POI state information corresponding to the POI name; and respectively performing POI name label prediction and POI state label prediction on the word sequence corresponding to the text by the POI state recognition model, and aligning the POI name label prediction result and the POI state label prediction result to obtain the binary group. The application can save the labor cost and improve the timeliness and the accuracy.

摘要：【中文】本申请公开了一种资源处理方法、装置以及电子设备，涉及资源处理领域。所述方法包括：确定与资源的类型对应的资源配置规则；选择与资源的配置规则对应的一个或多个算子，并获取各算子的运行顺序和运行状态；根据各算子的运行顺序和运行状态对资源进行处理，得到处理结果。无需针对不同类型的资源分别进行处理，减少了开发成本，有利于扩展新的资源，提高了资源处理效率。 【EN】The application discloses a resource processing method and device and electronic equipment, and relates to the field of resource processing. The method comprises the following steps: determining a resource configuration rule corresponding to the type of the resource; selecting one or more operators corresponding to the configuration rule of the resource, and acquiring the operation sequence and the operation state of each operator; and processing the resources according to the running sequence and the running state of each operator to obtain a processing result. The resources of different types do not need to be processed respectively, development cost is reduced, new resources can be expanded, and resource processing efficiency is improved.

摘要：【中文】本申请公开了一种韵律停顿等级预测的校正方法、装置、设备和介质，涉及韵律停顿等级预测技术。具体实现方案为：获取输入文本的韵律停顿等级的初始预测结果序列；获取所述初始预测结果序列的状态向量，其中，状态向量用于表示所述初始预测结果序列中各元素之间的依赖关系；将所述输入文本的语义向量与所述状态向量进行融合，得到融合向量；将所述融合向量输入条件随机场网络，得到校正后的预测结果序列。本申请实施例将语义向量与状态向量融合后再进行CRF，由于状态向量能表示预测结果序列中各元素之间的依赖关系，因此融合后不仅结合了语义信息，还结合了预测结果序列中各元素之间的依赖关系，从而提高CRF输出的最终校正后的预测结果序列的准确度。 【EN】The application discloses a method, a device, equipment and a medium for correcting prosody pause grade prediction, and relates to a prosody pause grade prediction technology. The specific implementation scheme is as follows: acquiring an initial prediction result sequence of a prosody pause grade of an input text; acquiring a state vector of the initial prediction result sequence, wherein the state vector is used for representing the dependency relationship among elements in the initial prediction result sequence; fusing the semantic vector of the input text with the state vector to obtain a fused vector; and inputting the fusion vector into a conditional random field network to obtain a corrected prediction result sequence. According to the embodiment of the application, the CRF is carried out after the semantic vector and the state vector are fused, and the state vector can represent the dependency relationship among elements in the prediction result sequence, so that not only is the semantic information combined after the fusion, but also the dependency relationship among the elements in the prediction result sequence is combined, and the accuracy of the final corrected prediction result sequence output by the CRF is improved.

摘要：【中文】本申请公开了一种面向领域的兴趣点选择方法、装置以及电子设备，涉及兴趣点选择领域。具体实现方案为：从兴趣点全集中选择目标兴趣点，并确定目标领域；获取包括多个样本的样本集合，各样本包括所属领域以及目标兴趣点；利用样本集合计算目标兴趣点与目标领域之间的相关度；根据相关度，得到目标领域相关的兴趣点。提高了兴趣点选择的适用性，能够更好的针对用户的兴趣进行兴趣点选择，能够适应样本规模不断增加的场景。 【EN】The application discloses a field-oriented interest point selection method and device and electronic equipment, and relates to the field of interest point selection. The specific implementation scheme is as follows: selecting a target interest point from the interest point complete set, and determining a target field; obtaining a sample set comprising a plurality of samples, wherein each sample comprises a field and a target interest point; calculating the correlation between the target interest point and the target field by utilizing the sample set; and obtaining the interest points related to the target field according to the correlation. The method and the device improve the applicability of interest point selection, can better select interest points according to the interest of the user, and can adapt to scenes with continuously increased sample scale.

摘要：【中文】本申请公开了视频直播的监控方法、装置、电子设备及存储介质，涉及视频直播领域。具体实现方案为：获取视频直播的直播间封面图像及直播间标题，并抽取所述视频直播中的多个实时帧图像；对所述直播间封面图像及所述实时帧图像的图像质量打分，得到图像质量分数；并对所述直播间标题进行敏感词匹配校验，得到敏感词匹配结果；在所述图像质量分数及所述敏感词匹配结果中的至少一项不满足预设要求的情况下，对所述视频直播进行干预。本申请实施例能够提高视频直播的监控效率和监控效果。 【EN】The application discloses a video live broadcast monitoring method and device, electronic equipment and a storage medium, and relates to the field of video live broadcast. The specific implementation scheme is as follows: acquiring a cover image and a title of a live broadcast room of a live video, and extracting a plurality of real-time frame images in the live video; scoring the image quality of the cover image of the live broadcast room and the real-time frame image to obtain an image quality score; performing sensitive word matching verification on the live broadcast room title to obtain a sensitive word matching result; and under the condition that at least one of the image quality score and the sensitive word matching result does not meet the preset requirement, intervening the live video. The embodiment of the application can improve the monitoring efficiency and the monitoring effect of live video.