会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 1. 发明申请
    • METHOD AND APPARATUS FOR GENERATING A SORTED LIST OF ITEMS
    • 用于产生分类项目列表的方法和装置
    • WO2013034554A1
    • 2013-03-14
    • PCT/EP2012/067228
    • 2012-09-04
    • AXEL SPRINGER DIGITAL TV GUIDE GMBHKORST, JanPRONK, Serverius, Petrus, PaulusBARBIERI, MauroCLOUT, Ramon
    • KORST, JanPRONK, Serverius, Petrus, PaulusBARBIERI, MauroCLOUT, Ramon
    • G06F17/30H04N21/482H04N21/466
    • G06F17/30554G06F17/30058G06F17/3053G06F17/3084H04N21/466H04N21/4668H04N21/4755H04N21/482H04N21/4826H04N21/8133H04N21/84
    • The invention relates to an electronic device and a method for automatic generation of a sorted list of items related to a seed-item. The electronic device comprises a relatedness determinator that is configured to compare said seed item with a plurality of further items and to thus determine a relatedness value for each further item with respect to said seed item. The device device further comprises a clustering engine that is configured to cluster the further items by determining a relative relatedness between (among) the further items. Thus, each further item is assigned to one cluster. The device further comprises a list generator that is configured to generate a sorted result list by sorting the further items according to both, their relatedness value and their belonging (or membership) to a cluster in that once an item is added to the sorted list, the relatedness value depending ranking of the further items in that cluster is at least momentary lowered so as to promote adding items of further clusters thus achieving a higher momentary relatedness value depending ranking to the top of the sorted list.
    • 本发明涉及一种电子装置和一种用于自动生成与种子物品有关的一系列物品的排序清单的方法。 电子设备包括相关性确定器,其被配置为将所述种子项目与多个其他项目进行比较,并且因此确定关于所述种子项目的每个其他项目的相关性值。 所述设备设备还包括聚类引擎,其经配置以通过确定所述其它项目之间的相对相关性来聚集所述其他项目。 因此,每个其他项目被分配给一个集群。 该设备还包括列表生成器,其被配置为通过根据将相关性值及其归属(或成员资格)两者分组到其他项目来生成排序结果列表,因为一旦将项目添加到排序列表中, 该集群中的其他项目的相关性值取决于排名至少被暂时降低,以便促进添加更多聚类的项目,从而根据排序列表顶部的排名实现更高的瞬时相关性值。
    • 2. 发明申请
    • METHOD AND APPARATUS FOR AUTOMATIC GENERATION OF RECOMMENDATIONS
    • 自动生成推荐方法和装置
    • WO2013034553A1
    • 2013-03-14
    • PCT/EP2012/067227
    • 2012-09-04
    • AXEL SPRINGER DIGITAL TV GUIDE GMBHKORST, JanBARBIERI, MauroCLOUT, RamonPRONK, Serverius, Petrus, Paulus
    • KORST, JanBARBIERI, MauroCLOUT, RamonPRONK, Serverius, Petrus, Paulus
    • G06F17/30
    • H04N21/4668G06F17/30867H04N21/4828
    • A recommender engine (100) for recommending content items to a user comprises: a profile generation unit having a pre-profile input unit (110), which is configured to receive pre-profile data comprising pre-profile text data suitable for identifying entities of interest to a given user, and having a pre-profile analysis unit (112), which is connected with the pre-profile input unit (110) and configured to extract from the pre-profile data identification data identifying the entities of interest and to generate an initial user profile data set for the given user from the extracted identification data; a query generation unit (120), which is connected with the profile generation unit and configured to generate, using the extracted identification data from the initial user profile data set, at least two queries semantically different from each other to be directed to at least one content repository; a content retrieval unit (128), which is connected with the query generation unit (120) and configured to issue the generated queries to the at least one content repository (122) and which is configured to receive from the at least one content repository, in response to the queries, content-related response data comprising respective hit lists having at least one respective content-location identifier indicative of a storage location of a respective content item; and an interleaver unit (132), which is connected with the content retrieval unit and which is configured to generate from the different hit lists a single recommendation list by interleaving the content-location identifiers comprised in different ones of the hit lists with each other.
    • 用于向用户推荐内容的推荐引擎(100)包括:简档生成单元,其具有预配置文件输入单元(110),其被配置为接收包括预先配置文件数据的预配置文件数据,所述预配置文件数据适合于识别 给定用户的兴趣,以及具有预配置分析单元(112),其与预配置文输入单元(110)连接并且被配置为从识别感兴趣的实体的预先简档数据识别数据中提取,并且 从提取的识别数据生成给定用户的初始用户简档数据集; 查询生成单元(120),其与所述简档生成单元连接并且被配置为使用所提取的来自所述初始用户简档数据集的标识数据生成至少两个语义上彼此不同的查询,以指向至少一个 内容存储库 内容检索单元(128),其与所述查询生成单元(120)连接并且被配置为向所述至少一个内容存储库(122)发布所生成的查询,并且被配置为从所述至少一个内容存储库接收, 响应于所述查询,内容相关响应数据包括具有指示相应内容项目的存储位置的至少一个相应内容位置标识符的相应命中列表; 以及交织单元,其与内容检索单元连接,并且被配置为通过将包含在不同命中列表中的内容位置标识符彼此交织来从不同命中列表生成单个推荐列表。
    • 7. 发明申请
    • PROCESSING A CONTENT SIGNAL
    • 处理内容信号
    • WO2009074773A1
    • 2009-06-18
    • PCT/GB2008/003999
    • 2008-12-04
    • AMBX UK LIMITEDPRONK, Serverius, Petrus, PaulusKORST, Johannes, Henricus, Maria
    • PRONK, Serverius, Petrus, PaulusKORST, Johannes, Henricus, Maria
    • G06F17/30H04H60/56
    • G06F17/30823H04H20/31H04H60/37H04H60/56H04H60/73
    • A method of processing a content signal comprises receiving a content signal comprising a series of frames, selecting a plurality of frames in the content signal according to a predefined algorithm, the selected frames comprising a start frame, an end frame, and only a portion of the frames between the start frame and the end frame, extracting a fingerprint for each selected frame to create a set of fingerprints, matching a fingerprint from the set of fingerprints to a plurality of stored fingerprints, and identifying from the plurality of matched fingerprints, a best match, the identifying comprising for further fingerprints in the set of fingerprints, computing an error rate, against corresponding stored fingerprints, and determining which of the plurality of matched fingerprints has the lowest error rate. Advantageously, the step of selecting the plurality of frames comprises selecting frames that are non-successive in the content signal.
    • 处理内容信号的方法包括接收包括一系列帧的内容信号,根据预定义的算法在内容信号中选择多个帧,所选择的帧包括起始帧,结束帧,并且仅一部分 起始帧和结束帧之间的帧,为每个所选择的帧提取指纹,以创建一组指纹,将指纹从指纹集匹配到多个存储的指纹,以及从多个匹配的指纹中识别, 所述识别包括用于指纹集中的进一步指纹,针对相应的存储的指纹计算错误率,以及确定多个匹配指纹中的哪一个具有最低错误率。 有利地,选择多个帧的步骤包括选择内容信号中不连续的帧。