WO2017000623A1 - 一种信息推荐方法和装置 - Google Patents
一种信息推荐方法和装置 Download PDFInfo
- Publication number
- WO2017000623A1 WO2017000623A1 PCT/CN2016/078986 CN2016078986W WO2017000623A1 WO 2017000623 A1 WO2017000623 A1 WO 2017000623A1 CN 2016078986 W CN2016078986 W CN 2016078986W WO 2017000623 A1 WO2017000623 A1 WO 2017000623A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- recommendation
- user
- information
- emotional state
- preset
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
Definitions
- This application relates to, but is not limited to, information processing techniques.
- the recommendation system of the related art only extracts the product information of the user's interest based on the user's historical record, and obviously cannot fully capture the user's interest intention, and does not fully utilize various available technologies to obtain the user's behavior and reaction information, and cannot It is a good way to determine the user's intentions.
- the present invention provides an information recommendation method and apparatus, which can provide recommendation data to a user based on the user's emotional state information, thereby capturing the user's needs more accurately.
- a method of information recommendation includes:
- the third recommendation information is displayed.
- the preset recommendation manner includes at least one of the following:
- collaborative filtering recommendation Based on content recommendation, collaborative filtering recommendation, rule-based recommendation, utility-based recommendation, and knowledge-based recommendation.
- the preset combination algorithm includes:
- Weighting transforming, blending, feature combination, cascading, feature expansion, or meta-level.
- the obtaining the emotional state information of the user includes:
- the emotional state parameter includes at least one of the following:
- Heart rate pulse, respiratory rate, body temperature, speech intensity, impedance value, acceleration.
- a computer readable storage medium storing computer executable instructions that, when executed, implement the information recommendation method described above.
- An information recommendation device includes:
- the obtaining unit is configured to: obtain the emotional state information of the user;
- a first generating unit configured to: generate first recommendation information according to the acquired emotional state information
- the second generating unit is configured to: generate second recommendation information according to the preset recommendation manner and the history information of the user;
- the processing unit is configured to: process the first recommendation information and the second recommendation information according to a preset combination algorithm to obtain third recommendation information;
- the display unit is configured to: display the third recommendation information.
- the preset recommendation manner includes at least one of the following:
- the preset combination algorithm includes:
- Weighting transforming, blending, feature combination, cascading, feature expansion, or meta-level.
- the obtaining unit is set to:
- the emotional state parameter includes at least one of the following:
- Heart rate pulse, respiratory rate, body temperature, speech intensity, impedance value, acceleration.
- the recommendation data can be provided to the user based on the emotional state information of the user, thereby capturing the user's needs more accurately.
- FIG. 1 is a schematic flowchart of a method for recommending information according to an embodiment of the present invention
- FIG. 2 is a schematic structural diagram of an apparatus configured to acquire emotional state information of a user according to an embodiment of the present invention
- FIG. 3 is a schematic structural diagram of an information recommendation apparatus according to an embodiment of the present invention.
- the basic idea of the embodiment of the present invention is that, in view of the advantages and disadvantages of the various recommended methods of the related art, in the technical solution of the embodiment of the present invention, based on the real-time emotional and emotional changes of the user and the emotional state change curve of the user history, Using one or more of the recommended methods of the related art to combine with the user's current real-time emotions and historical emotions, presenting the user with a more suitable recommendation knot. fruit.
- the embodiment of the invention provides a method for recommending information, as shown in FIG. 1 , the method includes:
- Step 101 Acquire emotional state information of the user.
- step 101 may include:
- the emotional state parameter may include at least one of the following: heart rate, pulse rate, respiratory rate, body temperature, voice intensity, impedance value, and acceleration.
- step 101 the method for obtaining the user's emotional state information in step 101 can be implemented by using the technical means of the related art.
- the embodiments of the present invention are not described herein, and are briefly introduced as follows:
- S101 is a user heart rate, pulse, and respiratory monitoring acquisition unit
- S102 is a user body temperature monitoring and collecting unit
- S103 is a user voice and sound wave monitoring and collecting unit
- S104 is a user impedance monitoring acquisition unit
- S105 is a user acceleration monitoring acquisition unit
- S111 is a user terminal central controller
- S121 is a user terminal emotion output and display unit.
- the user impedance monitoring and collecting unit S104 can measure the body impedance value by collecting the biological current of the human body, and the impedance value of the user can reflect the user's emotion to a certain extent. For example, when the emotional state is good, the user's impedance value is low, and the emotional state is poor. The user's impedance value is high; the user acceleration monitoring acquisition unit S105 can obtain the user's acceleration value, and the user's acceleration value can also reflect the user's emotion to a certain extent. For example, when the emotional state is good, the user's motion amount is larger, and the acceleration value is larger. When the emotional state is poor, the user has a relatively negative response to the motion. The acceleration value is small. It is worth mentioning that the results collected by the above five acquisition units can be aggregated and analyzed by the user terminal central controller S111 to obtain a more comprehensive and accurate user emotion. status information.
- Step 102 Generate first recommendation information according to the acquired emotional state information.
- the first or more sets of real-time emotional information (ie, one or more of the foregoing emotional state information) that are obtained in real time from the user terminal emotion output and display unit S121 may be stored.
- abnormal data cleaning is performed on the captured source data, and data with large mood fluctuations is filtered and processed, so that the processed data can be used more effectively in subsequent operations;
- the user emotion information obtained by the foregoing processing is converted into a unit score of the unit according to the custom emotion conversion table;
- the emotion conversion table is generally customized by each business system; for example, the implementation of the custom emotion conversion table is shown in Table 1.
- the generated first recommendation information may be: light music, inspirational movie, taste stimulating food, yoga.
- step 101 after the user's current emotional state information is acquired, the device may be stored in the device.
- the historical emotional state information of the user is comprehensively analyzed, and the result is provided to step 102.
- step 102 the user's emotional state information obtained by comprehensively analyzing the step 101 can generate more accurate recommendation information for the user.
- Step 103 Generate second recommendation information according to the preset recommendation manner and the history information of the user.
- the preset recommendation manner may include at least one of the following:
- collaborative filtering recommendation Based on content recommendation, collaborative filtering recommendation, rule-based recommendation, utility-based recommendation, and knowledge-based recommendation.
- Step 104 Process the first recommendation information and the second recommendation information according to a preset combination algorithm to obtain third recommendation information.
- the preset combination algorithm may include: weighting, transforming, mixing, feature combination, layering, feature expansion, or meta-level.
- Weight Weights the recommendation results of a plurality of recommended techniques to obtain recommended results to the user.
- a rough recommendation result is generated by a recommendation technique, and another recommendation technique is used to further make a more accurate recommendation based on the recommendation result.
- Feature augmentation A technique that generates additional feature information embedded into the feature input of another recommendation technique.
- Meta-level A model generated by a recommended method/method is used as an input to another recommended method/method.
- the recommendation scheme of the embodiment of the present invention is related to the related art. There is a better recommendation effect, because the additional user emotions are added to make the final recommendation more in line with user expectations. Mainly reflected in the following three aspects: First, improve the current psychological and emotional state of the recommendation, such as, recommend music, movies, food that is more suitable for the current psychological state; Second, cultivate behavioral habits, hobbies, shape personality, such as, combined Long-term emotional state and current emotions recommend suitable books, sports, colors, social circles, etc. Third, combined with long-term emotional state and current emotional state, give a great contrast recommendation, break through the self, and achieve improvement in long-term or short-term The negative psychological state.
- Step 105 Display the third recommendation information.
- the third recommendation information finally generated may be displayed to the user.
- the device returns to step 103 to regenerate new recommendation information for the user.
- the recommendation data can be provided to the user based on the emotional state information of the user, thereby capturing the user's needs more accurately.
- the embodiment of the invention further provides a computer readable storage medium storing computer executable instructions, which are implemented when the computer executable instructions are executed.
- An embodiment of the present invention provides an information recommendation apparatus 10. As shown in FIG. 3, the apparatus 10 includes:
- the obtaining unit 11 is configured to: acquire emotional state information of the user;
- the first generating unit 12 is configured to: generate first recommendation information according to the acquired emotional state information
- the second generating unit 13 is configured to: generate second recommendation information according to the preset recommendation manner and the history information of the user;
- the processing unit 14 is configured to: process the first recommendation information and the second recommendation information according to a preset combination algorithm to obtain third recommendation information;
- the display unit 15 is configured to: display the third recommendation information.
- the preset recommendation manner includes at least one of the following:
- collaborative filtering recommendation Based on content recommendation, collaborative filtering recommendation, rule-based recommendation, utility-based recommendation, and knowledge-based recommendation.
- the preset combination algorithm includes:
- Weighting transforming, blending, feature combination, cascading, feature expansion, or meta-level.
- the obtaining unit 11 is configured to:
- the emotional state parameter includes at least one of the following:
- Heart rate pulse, respiratory rate, body temperature, speech intensity, impedance value, acceleration.
- the first generating unit 12 may include:
- the source data real-time capture storage unit 120 is configured to: store one or more sets of real-time emotional information (ie, one or more of the foregoing emotional state information) acquired from the user terminal emotion output and display unit S121 in real time. Go to the storage medium to provide raw data for subsequent processing operations;
- the source data cleaning, filtering, and processing unit 121 is configured to: perform abnormal data cleaning on the source data stored in the real-time data storage unit 120 of the source data, and filter and process the data with large mood fluctuations, so that the processed data can be processed. More effectively used in later operations;
- the data conversion unit 122 is configured to: convert the user emotion information obtained by the source data cleaning, filtering, and processing unit 121 into a unified unit of emotion score according to the custom emotion conversion table;
- the user portrait unit 123 is configured to: use the user sentiment score converted by the data conversion unit 122, and combine the weight ratio corresponding to the emotion score, and finally calculate the weighted average of all the emotion scores, that is, the current real time of the user. Emotional score
- the matching recommendation unit 124 is configured to compare and compare the emotion score obtained above with the “commodity emotion” score, and then give the recommendation list in priority order.
- the recommendation data can be provided to the user based on the emotional state information of the user, thereby capturing the user's needs more accurately.
- all or part of the steps of the above embodiments may also be implemented by using an integrated circuit. These steps may be separately fabricated into individual integrated circuit modules, or multiple modules or steps may be fabricated into a single integrated circuit module. achieve.
- the devices/function modules/functional units in the above embodiments may be implemented by a general-purpose computing device, which may be centralized on a single computing device or distributed over a network of multiple computing devices.
- the device/function module/functional unit in the above embodiment When the device/function module/functional unit in the above embodiment is implemented in the form of a software function module and sold or used as a stand-alone product, it can be stored in a computer readable storage medium.
- the above mentioned computer readable storage medium may be a read only memory, a magnetic disk or an optical disk or the like.
- the recommendation information is generated and displayed based on the emotional state information of the user. Therefore, the embodiment of the present invention can provide recommendation data for the user, thereby capturing the user's needs more accurately.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
提供了一种信息推荐方法和装置。所述方法包括：获取用户的情绪状态信息；根据获取的所述情绪状态信息生成第一推荐信息；根据预设推荐方式以及所述用户的历史记录信息生成第二推荐信息；根据预设组合算法对所述第一推荐信息与所述第二推荐信息进行处理，得到第三推荐信息；显示所述第三推荐信息。
Description
本申请涉及但不限于信息处理技术。
移动互联网时代，网上购物已经越来越方便、快捷，成为很多用户购物的首选方式，甚至是购买任何东西都先到网上查一查相关的资料信息，做到胸中有数，然而在纷繁芜杂的网上商店中要找到哪一款让你称心如意的商品也必然要耗费很多精力和时间。在现阶段，各种网上购物商场、系统都有类似的商品推荐系统，推荐系统通常基于用户或用户相关人群的历史数据来作为源数据分析并得出最终的推荐列表，将推荐列表中的产品信息提供给用户。
发明内容
以下是对本文详细描述的主题的概述。本概述并非是为了限制权利要求的保护范围。
相关技术的推荐系统仅根据用户的历史记录来挖掘用户感兴趣的产品信息显然不能够全面地捕捉用户的兴趣意向，也没有充分地利用各种可用的技术来获取用户的行为和反应信息，不能很好地确定用户的需求意向。
本文提供一种信息推荐方法和装置，能够基于用户的情绪状态信息为用户提供推荐数据，从而较准确地捕捉用户的需求。
一种信息推荐方法包括：
获取用户的情绪状态信息；
根据获取的所述情绪状态信息生成第一推荐信息；
根据预设推荐方式以及所述用户的历史记录信息生成第二推荐信息；
根据预设组合算法对所述第一推荐信息与所述第二推荐信息进行处理，
得到第三推荐信息；
显示所述第三推荐信息。
可选地，所述预设推荐方式包括以下至少一种：
基于内容推荐、协同过滤推荐、基于规则推荐、基于效用推荐、基于知识推荐。
可选地，所述预设组合算法包括：
加权、变换、混合、特征组合、层叠、特征扩充或元级别。
可选地，所述获取用户的情绪状态信息包括：
采集所述用户当前的情绪状态参数；
根据所述用户当前的情绪状态参数以及所述用户的历史的情绪状态参数获取所述用户的情绪状态信息。
可选地，所述情绪状态参数包括以下至少一种：
心率、脉搏、呼吸频率、体温、语音强度、阻抗值、加速度。
一种计算机可读存储介质，存储有计算机可执行指令，所述计算机可执行指令被执行时实现上述信息推荐方法。
一种信息推荐装置包括：
获取单元，设置为：获取用户的情绪状态信息；
第一生成单元，设置为：根据获取的所述情绪状态信息生成第一推荐信息；
第二生成单元，设置为：根据预设推荐方式以及所述用户的历史记录信息生成第二推荐信息；
处理单元，设置为：根据预设组合算法对所述第一推荐信息与所述第二推荐信息进行处理，得到第三推荐信息；
显示单元，设置为：显示所述第三推荐信息。
可选地，所述预设推荐方式包括以下至少一种：
基于内容推荐、协同过滤推荐、基于规则推荐、基于效用推荐、基于知
识推荐。
可选地，所述预设组合算法包括：
加权、变换、混合、特征组合、层叠、特征扩充或元级别。
可选地，所述获取单元是设置为：：
采集所述用户当前的情绪状态参数；
根据所述用户当前的情绪状态参数以及所述用户的历史的情绪状态参数获取所述用户的情绪状态信息。
可选地，所述情绪状态参数包括以下至少一种：
心率、脉搏、呼吸频率、体温、语音强度、阻抗值、加速度。
通过本发明实施例的方案，能够基于用户的情绪状态信息为用户提供推荐数据，从而较准确地捕捉用户的需求。
在阅读并理解了附图和详细描述后，可以明白其他方面。
附图概述
图1为本发明实施例提供的一种信息推荐方法的流程示意图；
图2为本发明实施例提供的一种设置为获取用户的情绪状态信息的装置的结构示意图；
图3为本发明实施例提供的一种信息推荐装置的结构示意图。
本发明的较佳实施方式
下面结合附图对本发明的实施方式进行描述。需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的各种方式可以相互组合。
本发明实施例的基本思想是：由于相关技术的各种推荐方式都有优缺点，在本发明实施例的技术方案中，基于用户实时的情绪、情感变化以及该用户历史的情绪状态变化曲线，利用相关技术的推荐方式中的一种或多种组合与用户当前实时情绪与历史情绪相结合，给用户当下展示更适合的推荐结
果。
本发明实施例提供了一种信息推荐方法，如图1所示，该方法包括：
步骤101、获取用户的情绪状态信息。
其中，步骤101可以包括：
采集所述用户当前的情绪状态参数；
根据所述用户当前的情绪状态参数以及所述用户的历史的情绪状态参数获取所述用户的情绪状态信息；
其中，所述情绪状态参数可包括以下至少一种：心率、脉搏、呼吸频率、体温、语音强度、阻抗值、加速度。
需要说明的是，对于步骤101所述的获取用户的情绪状态信息可以采用相关技术的技术手段实现，本发明实施例在此不做赘述，示意性地简要介绍如下：
如图2所示的一种设置为获取用户的情绪状态信息的装置中，S101为用户心率、脉搏、呼吸监测采集单元；S102为用户体温监测采集单元；S103为用户语音、声波监测采集单元；S104为用户阻抗监测采集单元；S105为用户加速度监测采集单元；S111为用户终端中央控制器；S121为用户终端情绪输出、展示单元。
其中，用户阻抗监测采集单元S104可以通过采集人体的生物电流来测算人体阻抗值，用户的阻抗值一定程度上可以反映用户的情绪，例如在情绪状态良好时用户的阻抗值较低，情绪状态差时用户的阻抗值较高；用户加速度监测采集单元S105可以获取用户的加速度值，用户的加速度值一定程度上也可以反映用户的情绪，例如，情绪状态良好时用户运动量较大对应加速度值较大，情绪状态差时用户对运动比较消极对应加速度值较小；值得一提的是，上述5个采集单元采集到的结果可以由用户终端中央控制器S111来汇总分析得到更加全面、准确的用户情绪状态信息。
步骤102、根据获取的所述情绪状态信息生成第一推荐信息。
其中，可以首先将从上述用户终端情绪输出、展示单元S121实时获取的一组或多组用户实时情绪信息(即前述情绪状态信息中的一种或多种)存
储到存储介质中，为后续处理操作提供原始数据；
然后，对抓取到的源数据进行异常数据清洗，情绪波动较大的数据进行过滤并处理，使得处理过后的数据能更有效地在后面的操作中使用；
然后，将前述处理后得到的用户情绪信息按照自定义情绪转换表转换为统一单位的情绪分值；
由于不同的业务系统对情绪的范围定义可能存在偏差，该情绪转换表一般由每个业务系统自定义；示例性地，自定义情绪转换表的实施范例如表1所示。
表1
| 心率 | 40次/分 | 60次/分 | 80次/分 | 100次/分 | 120次/分 |
| 情绪分值 | 60 | 100 | 120 | 160 | 200 |
然后，利用前述转换得到的一组或多组用户情绪分值，并结合该情绪分值所对应的权重比值，最后计算出所有情绪分值的加权平均数，即为用户当前实时情绪分值；
最后，将上述得到的情绪分值与“商品情绪”分值进行比较对比，然后，按照优先顺序给出推荐列表。
示例性地，如表2所示为推荐信息生成的实施范例。
表2
例如，当用户当前实时情绪分值为20时，生成的第一推荐信息可以是：轻音乐、励志电影、味蕾刺激强的食物、瑜伽。
需要说明的是，为了提高根据用户情绪状态信息为用户推荐信息的准确性，步骤101中，可以在获取用户当前的情绪状态信息后与该装置中存储的
用户的历史情绪状态信息进行综合分析后将结果提供给步骤102，步骤102中则可以根据步骤101综合分析后得到的用户的情绪状态信息为用户生成更为准确的推荐信息。
步骤103、根据预设推荐方式以及所述用户的历史记录信息生成第二推荐信息。
其中，所述预设推荐方式可包括以下至少一种：
基于内容推荐、协同过滤推荐、基于规则推荐、基于效用推荐、基于知识推荐。
需要说明的是，上述五种推荐方式均属于现有技术方案中的推荐方式。本发明实施例对上述每一种推荐方式的具体方案不做详细阐述。在本发明实施例中仅需要利用上述的推荐方式获得第二推荐信息。
步骤104、根据预设组合算法对所述第一推荐信息与所述第二推荐信息进行处理，得到第三推荐信息。
其中，所述预设组合算法可包括：加权、变换、混合、特征组合、层叠、特征扩充或元级别。
上述7种组合算法属于相关技术的推荐系统中的算法，简单介绍如下：
1)加权(Weight)：加权多种推荐技术的推荐结果，得到向用户提供的推荐结果。
2)变换(Switch)：根据问题背景和实际情况或要求，决定采用不同的一种或多种推荐技术进行变换。
3)混合(Mixed)：同时采用多种推荐技术、给出多种推荐结果，为用户提供参考。
4)特征组合(Feature combination)：组合来自不同推荐数据源的特征被另一种推荐算法所采用。
5)层叠(Cascade)：先用一种推荐技术产生一种粗糙的推荐结果，再用另一种推荐技术在此推荐结果的基础上进一步作出更精确的推荐。
6)特征扩充(Feature augmentation)：一种技术产生附加的特征信息嵌入到另一种推荐技术的特征输入中。
7)元级别(Meta-level)：用一种推荐方式/方法产生的模型作为另一种推荐方式/方法的输入。
需要特别说明的是，由于本发明实施例中利用预设组合算法生成第三推荐信息时使用了根据用户情绪状态信息生成的第一推荐信息，因此，本发明实施例相对于相关技术的推荐方案有更好的推荐效果，由于额外增加了用户情绪这一维度，使最终的推荐结果更加符合用户预期。主要体现在以下三个方面：一、改善当前心理、情绪状态的推荐，诸如，推荐更适合当下心理状态的音乐、电影、美食；二、培养行为习惯、兴趣爱好，塑造个性性格，诸如，结合长期的情绪状态与当前的情绪推荐适合的书籍、运动、颜色、社交圈子等；三、结合长期的情绪状态与当下的情绪状态，给出反差极大的推荐，突破自我，达到改善长期或短期的负面心理状态。
步骤105、显示所述第三推荐信息。
其中，可以向用户显示最终生成的第三推荐信息。
另外，值得一提的是，如果用户对推荐的信息不满意,向该装置反馈后，该装置则返回步骤103重新为用户生成新的推荐信息。
通过本发明实施例的方案，能够基于用户的情绪状态信息为用户提供推荐数据，从而较准确地捕捉用户的需求。
虽然本申请的流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤。
本发明实施例还提供了一种计算机可读存储介质，存储有计算机可执行指令，所述计算机可执行指令被执行时实现上述信息推荐方法。
本发明实施例提供了一种信息推荐装置10，如图3所示，该装置10包括：
获取单元11，设置为：获取用户的情绪状态信息；
第一生成单元12，设置为：根据获取的所述情绪状态信息生成第一推荐信息；
第二生成单元13，设置为：根据预设推荐方式以及所述用户的历史记录信息生成第二推荐信息；
处理单元14，设置为：根据预设组合算法对所述第一推荐信息与所述第二推荐信息进行处理，得到第三推荐信息；
显示单元15，设置为：显示所述第三推荐信息。
可选地，所述预设推荐方式包括以下至少一种：
基于内容推荐、协同过滤推荐、基于规则推荐、基于效用推荐、基于知识推荐。
可选地，所述预设组合算法包括：
加权、变换、混合、特征组合、层叠、特征扩充或元级别。
可选地，所述获取单元11是设置为：
采集所述用户当前的情绪状态参数；
根据所述用户当前的情绪状态参数以及所述用户的历史的情绪状态参数获取所述用户的情绪状态信息。
可选地，所述情绪状态参数包括以下至少一种：
心率、脉搏、呼吸频率、体温、语音强度、阻抗值、加速度。
其中，所述第一生成单元12可以包括：
源数据实时抓取存储单元120、源数据清洗、过滤及处理单元121、数据转换单元122、用户画像单元123以及匹配推荐单元124；
源数据实时抓取存储单元120，设置为：将从上述用户终端情绪输出、展示单元S121实时获取的一组或多组用户实时情绪信息(即前述情绪状态信息中的一种或多种)存储到到存储介质中，为后续处理操作提供原始数据；
源数据清洗、过滤及处理单元121，设置为：对源数据实时抓取存储单元120中存储的源数据进行异常数据清洗，对情绪波动较大的数据进行过滤并处理，使得处理过后的数据能更有效地在后面的操作中使用；
数据转换单元122，设置为：将源数据清洗、过滤及处理单元121处理后得到的用户情绪信息按照自定义情绪转换表转换为统一单位的情绪分值；
用户画像单元123，设置为：利用数据转换单元122转换得到的用户情绪分值，并结合该情绪分值所对应的权重比值，最后计算出所有情绪分值的加权平均数，即为用户当前实时情绪分值；
匹配推荐单元124，设置为：将上述得到的情绪分值与“商品情绪”分值进行比较对比，然后，按照优先顺序给出推荐列表。
本实施例用于实现上述每个方法实施例，本实施例中每个单元的工作流程和工作原理参见上述每个方法实施例中的描述，在此不再赘述。
通过本发明实施例的方案，能够基于用户的情绪状态信息为用户提供推荐数据，从而较准确地捕捉用户的需求。
本领域普通技术人员可以理解上述实施例的全部或部分步骤可以使用计算机程序流程来实现，所述计算机程序可以存储于一计算机可读存储介质中，所述计算机程序在相应的硬件平台上(如系统、设备、装置、器件等)执行，在执行时，包括方法实施例的步骤之一或其组合。
可选地，上述实施例的全部或部分步骤也可以使用集成电路来实现，这些步骤可以被分别制作成一个个集成电路模块，或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。
上述实施例中的装置/功能模块/功能单元可以采用通用的计算装置来实现，它们可以集中在单个的计算装置上，也可以分布在多个计算装置所组成的网络上。
上述实施例中的装置/功能模块/功能单元以软件功能模块的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。上述提到的计算机可读取存储介质可以是只读存储器，磁盘或光盘等。
通过本发明实施例的方案，基于用户的情绪状态信息生成并显示推荐信息。因此，本发明实施例能够为用户提供推荐数据，从而较准确地捕捉用户的需求。
Claims (10)
- 一种信息推荐方法，包括：获取用户的情绪状态信息；根据获取的所述情绪状态信息生成第一推荐信息；根据预设推荐方式以及所述用户的历史记录信息生成第二推荐信息；根据预设组合算法对所述第一推荐信息与所述第二推荐信息进行处理，得到第三推荐信息；显示所述第三推荐信息。
- 根据权利要求1所述的方法，其中，所述预设推荐方式包括以下至少一种：基于内容推荐、协同过滤推荐、基于规则推荐、基于效用推荐、基于知识推荐。
- 根据权利要求1所述的方法，其中，所述预设组合算法包括：加权、变换、混合、特征组合、层叠、特征扩充或元级别。
- 根据权利要求1所述的方法，其中，所述获取用户的情绪状态信息包括：采集所述用户当前的情绪状态参数；根据所述用户当前的情绪状态参数以及所述用户的历史的情绪状态参数获取所述用户的情绪状态信息。
- 根据权利要求4所述的方法，其中，所述情绪状态参数包括以下至少一种：心率、脉搏、呼吸频率、体温、语音强度、阻抗值、加速度。
- 一种信息推荐装置，包括：获取单元，设置为：获取用户的情绪状态信息；第一生成单元，设置为：根据获取的所述情绪状态信息生成第一推荐信息；第二生成单元，设置为：根据预设推荐方式以及所述用户的历史记录信息生成第二推荐信息；处理单元，设置为：根据预设组合算法对所述第一推荐信息与所述第二推荐信息进行处理，得到第三推荐信息；显示单元，设置为：显示所述第三推荐信息。
- 根据权利要求6所述的装置，其中，所述预设推荐方式包括以下至少一种：基于内容推荐、协同过滤推荐、基于规则推荐、基于效用推荐、基于知识推荐。
- 根据权利要求6所述的装置，其中，所述预设组合算法包括：加权、变换、混合、特征组合、层叠、特征扩充或元级别。
- 根据权利要求6所述的装置，其中，所述获取单元是设置为：：采集所述用户当前的情绪状态参数；根据所述用户当前的情绪状态参数以及所述用户的历史的情绪状态参数获取所述用户的情绪状态信息。
- 根据权利要求9所述的装置，其中，所述情绪状态参数包括以下至少一种：心率、脉搏、呼吸频率、体温、语音强度、阻抗值、加速度。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510383481.3A CN106327234A (zh) | 2015-07-02 | 2015-07-02 | 一种信息推荐方法和装置 |
| CN201510383481.3 | 2015-07-02 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2017000623A1 true WO2017000623A1 (zh) | 2017-01-05 |
Family
ID=57607793
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2016/078986 WO2017000623A1 (zh) | 2015-07-02 | 2016-04-11 | 一种信息推荐方法和装置 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN106327234A (zh) |
| WO (1) | WO2017000623A1 (zh) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110110321A (zh) * | 2019-03-19 | 2019-08-09 | 深圳壹账通智能科技有限公司 | 基于语音数据的产品推荐方法、装置、设备及存储介质 |
| CN110781377A (zh) * | 2019-09-03 | 2020-02-11 | 腾讯科技（深圳）有限公司 | 一种文章推荐方法、装置 |
| CN111062785A (zh) * | 2019-12-18 | 2020-04-24 | 上海良鑫网络科技有限公司 | 智能选择产品来推荐给相匹配用户的方法与系统 |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107330747B (zh) * | 2017-05-16 | 2020-12-01 | 深圳和而泰智能家居科技有限公司 | 美容设备档位推荐方法、美容设备以及存储介质 |
| CN107666518B (zh) * | 2017-09-27 | 2023-03-03 | 百度在线网络技术（北京）有限公司 | 信息推送方法、装置、终端设备和计算机可读存储介质 |
| CN108446635B (zh) * | 2018-03-19 | 2022-03-22 | 西北大学 | 一种利用脑电信号辅助偏好获取协同过滤推荐系统及方法 |
| CN108876541A (zh) * | 2018-06-20 | 2018-11-23 | 郑州村村联网络技术有限公司 | 商品推荐方法及装置 |
| CN109543132A (zh) * | 2018-11-22 | 2019-03-29 | 深圳墨世科技有限公司 | 内容推荐方法、装置、电子设备及存储介质 |
| CN109740017A (zh) * | 2018-12-14 | 2019-05-10 | 北京字节跳动网络技术有限公司 | 推荐信息的即时反馈方法及装置 |
| CN110866146B (zh) * | 2019-11-26 | 2022-11-11 | 杭州市第一人民医院 | 视频推荐方法、装置、计算机设备和存储介质 |
| TWI841847B (zh) * | 2021-06-01 | 2024-05-11 | 中華電信股份有限公司 | 互動式情緒轉移方法與系統 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1395798A (zh) * | 2000-11-22 | 2003-02-05 | 皇家菲利浦电子有限公司 | 根据用户的当前心情产生推荐的方法和装置 |
| US20130283303A1 (en) * | 2012-04-23 | 2013-10-24 | Electronics And Telecommunications Research Institute | Apparatus and method for recommending content based on user's emotion |
| CN103559197A (zh) * | 2013-09-23 | 2014-02-05 | 浙江大学 | 基于上下文预过滤的实时音乐推荐方法 |
| CN103970873A (zh) * | 2014-05-14 | 2014-08-06 | 中国联合网络通信集团有限公司 | 一种音乐推荐方法和系统 |
| CN104202718A (zh) * | 2014-08-05 | 2014-12-10 | 百度在线网络技术（北京）有限公司 | 一种向用户提供信息的方法与装置 |
-
2015
- 2015-07-02 CN CN201510383481.3A patent/CN106327234A/zh active Pending
-
2016
- 2016-04-11 WO PCT/CN2016/078986 patent/WO2017000623A1/zh active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1395798A (zh) * | 2000-11-22 | 2003-02-05 | 皇家菲利浦电子有限公司 | 根据用户的当前心情产生推荐的方法和装置 |
| US20130283303A1 (en) * | 2012-04-23 | 2013-10-24 | Electronics And Telecommunications Research Institute | Apparatus and method for recommending content based on user's emotion |
| CN103559197A (zh) * | 2013-09-23 | 2014-02-05 | 浙江大学 | 基于上下文预过滤的实时音乐推荐方法 |
| CN103970873A (zh) * | 2014-05-14 | 2014-08-06 | 中国联合网络通信集团有限公司 | 一种音乐推荐方法和系统 |
| CN104202718A (zh) * | 2014-08-05 | 2014-12-10 | 百度在线网络技术（北京）有限公司 | 一种向用户提供信息的方法与装置 |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110110321A (zh) * | 2019-03-19 | 2019-08-09 | 深圳壹账通智能科技有限公司 | 基于语音数据的产品推荐方法、装置、设备及存储介质 |
| CN110781377A (zh) * | 2019-09-03 | 2020-02-11 | 腾讯科技（深圳）有限公司 | 一种文章推荐方法、装置 |
| CN110781377B (zh) * | 2019-09-03 | 2024-02-20 | 深圳市雅阅科技有限公司 | 一种文章推荐方法、装置 |
| CN111062785A (zh) * | 2019-12-18 | 2020-04-24 | 上海良鑫网络科技有限公司 | 智能选择产品来推荐给相匹配用户的方法与系统 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106327234A (zh) | 2017-01-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2017000623A1 (zh) | 一种信息推荐方法和装置 | |
| Ip et al. | Revisiting the antecedents of social entrepreneurial intentions in Hong Kong | |
| Herranz-Pascual et al. | Going beyond quietness: Determining the emotionally restorative effect of acoustic environments in urban open public spaces | |
| Li et al. | Human emotion recognition with electroencephalographic multidimensional features by hybrid deep neural networks | |
| Kim et al. | Towards emotionally aware AI smart classroom: Current issues and directions for engineering and education | |
| Mohino-Herranz et al. | Assessment of mental, emotional and physical stress through analysis of physiological signals using smartphones | |
| US10565213B2 (en) | Methods and systems for identifying similar schools | |
| Bordel et al. | Assessment of human motivation through analysis of physiological and emotional signals in Industry 4.0 scenarios | |
| Herranz-Pascual et al. | Analysis of field data to describe the effect of context (Acoustic and Non-Acoustic Factors) on urban soundscapes | |
| Choi et al. | Influence of SNS user innovativeness and public individuation on SNS usage patterns and social capital development: The case of Facebook | |
| Arnrich | A survey on measuring happiness with smart phones | |
| Othman et al. | Predicting group contribution behaviour in a public goods game from face-to-face communication | |
| Navarro-Alamán et al. | EmotIoT: an IoT system to improve users’ wellbeing | |
| Ciordas-Hertel et al. | Mobile sensing with smart wearables of the physical context of distance learning students to consider its effects on learning | |
| Burrows et al. | Realtime emotional reflective user interface based on deep convolutional neural networks and generative adversarial networks | |
| Liao et al. | Manipulating users’ trust of autonomous products with affective priming | |
| Wolk et al. | The therapeutic aspects of embroidery in art therapy from the perspective of adolescent girls in a post-hospitalization boarding school | |
| Correia et al. | Towards automatic image enhancement with genetic programming and machine learning | |
| Ottoni et al. | A deep learning approach for speech emotion recognition optimization using meta-learning | |
| US20180308180A1 (en) | Systems Methods Devices Circuits and Computer Executable Code for Impression Measurement and Evaluation | |
| Gilroy et al. | An affective model of user experience for interactive art | |
| Cuomo et al. | A biologically inspired model for analyzing behaviours in social network community and cultural heritage scenario | |
| Chacón et al. | Machine learning models and statistical complexity to analyze the effects of posture on cerebral hemodynamics | |
| KR20170050325A (ko) | 자유로운 공감 표현을 위하여 성장 이미지를 이용하는 소셜 네트워크 서비스 방법, 이를 위한 사용자 단말 및 서버 | |
| Pedroza et al. | Acoustic individual identification in birds based on the band-limited phase-only correlation function |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16816988 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 16816988 Country of ref document: EP Kind code of ref document: A1 |