Abstract: An information processing device includes a computation unit that computes a similarity between character strings among plural character strings which express a content of electronic information and which are generated from the electronic information, and an association unit that associates a dissimilar character string that is not similar to another character string among the plural character strings with the electronic information as a representative character string that expresses the electronic information.

Abstract: A non-transitory computer readable medium stores a program causing a computer to execute a process for learning. The process includes: generating, from an input text, an output text related to content of the input text and different from the input text by using a generation model that generates the output text from the input text; reconstructing the input text from the output text by using a reconstruction model that reconstructs the input text from the output text; and updating at least one of the generation model and the reconstruction model by causing the at least one of the generation model and the reconstruction model to perform learning by using a difference between the input text and a reconstructed text reconstructed in the reconstructing.

Abstract: A purchasing behavior analysis apparatus includes an acquiring unit that acquires posting information about a specific product from posting information posted to a social networking service; a conversion unit; and an artificial neural network. The artificial neural network includes a first determination layer which determines whether a user is interested in the product, a second determination layer which determines whether the user wants the product, and a third determination layer which determines whether the user is predicted to purchase the product in the future. The purchasing behavior analysis apparatus further includes an interest presence probability calculating unit, a purchase desire probability calculating unit, and a purchase likelihood probability calculating unit.

Abstract: An information processing device includes a computation unit that computes a similarity between character strings among plural character strings which express a content of electronic information and which are generated from the electronic information, and an association unit that associates a dissimilar character string that is not similar to another character string among the plural character strings with the electronic information as a representative character string that expresses the electronic information.

Abstract: A purchasing behavior analysis apparatus includes an acquiring unit that acquires posting information about a specific product from posting information posted to a social networking service; a conversion unit; and an artificial neural network. The artificial neural network includes a first determination layer which determines whether a user is interested in the product, a second determination layer which determines whether the user wants the product, and a third determination layer which determines whether the user is predicted to purchase the product in the future. The purchasing behavior analysis apparatus further includes an interest presence probability calculating unit, a purchase desire probability calculating unit, and a purchase likelihood probability calculating unit.

Abstract: An information processing apparatus includes a first extracting unit, a second extracting unit, and a third extracting unit. The first extracting unit applies a topic model to target text information and extracts topic distributions for words constituting the text information. The second extracting unit extracts a first topic for the text information from the topic distributions extracted by the first extracting unit. The third extracting unit extracts a word satisfying a predetermined condition, from at least one word having the first topic, as a context word in the text information. The first topic is extracted by the second extracting unit.

Abstract: A non-transitory computer readable medium storing a program is provided. A process of the program includes: extracting, concerning first document information including texts, a word sense associated with a word in each text; setting a word sense to be a teacher signal; creating a first topic model by using the teacher signal; estimating a topic of each text by using the first topic model; generating a learning model by using the topic as a feature and the teacher signal as a category; extracting, concerning second document information including texts, a word sense associated with a word in each text; setting a word sense to be a teacher signal; creating a second topic model by using the teacher signal; estimating a topic of each text by using the second topic model; and estimating a category of each text by using the topic as a feature and the learning model.

Abstract: An information processing apparatus includes a receiving unit, a determining unit, a first assigning unit, an extracting unit, a second assigning unit, a modeling unit, and an output unit. The receiving unit receives a target character string. The determining unit determines whether a sentiment character string is included in the received character string. A first assigning unit assigns a label corresponding to a sentiment character string to the character string when the sentiment character string is included, and assigns plural labels to the character string when no sentiment character string is included. The extracting unit extracts a word from the character string. The second assigning unit assigns to the extracted word a label which has been assigned to the character string that includes the word. The modeling unit performs supervised topic modeling for the character string. The output unit outputs a result of a process by the modeling unit.

Abstract: A non-transitory computer readable medium stores an annotation-information adding program that causes a computer to function as an adding unit, an evaluating unit, and a setting unit. The adding unit adds annotation information to target information including multiple targets based on input from a first inputter. The evaluating unit evaluates reliability of the first inputter and reliability of a second inputter by comparing annotation information already added to at least one of the multiple targets by the second inputter with annotation information added by the first inputter. The setting unit sets a target range in the target information intended for requesting the first inputter to add annotation information based on the reliability of the first inputter and the reliability of the second inputter.

Abstract: A non-transitory computer readable medium storing a program is provided. A process of the program includes: extracting, concerning first document information including texts, a word sense associated with a word in each text; setting a word sense to be a teacher signal; creating a first topic model by using the teacher signal; estimating a topic of each text by using the first topic model; generating a learning model by using the topic as a feature and the teacher signal as a category; extracting, concerning second document information including texts, a word sense associated with a word in each text; setting a word sense to be a teacher signal; creating a second topic model by using the teacher signal; estimating a topic of each text by using the second topic model; and estimating a category of each text by using the topic as a feature and the learning model.

Abstract: An information processing apparatus includes a receiving unit, a determining unit, a first assigning unit, an extracting unit, a second assigning unit, a modeling unit, and an output unit. The receiving unit receives a target character string. The determining unit determines whether a sentiment character string is included in the received character string. A first assigning unit assigns a label corresponding to a sentiment character string to the character string when the sentiment character string is included, and assigns plural labels to the character string when no sentiment character string is included. The extracting unit extracts a word from the character string. The second assigning unit assigns to the extracted word a label which has been assigned to the character string that includes the word. The modeling unit performs supervised topic modeling for the character string. The output unit outputs a result of a process by the modeling unit.

Abstract: There is provided a non-transitory computer-readable medium storing a program causing a computer to execute a process. The process includes: acquiring posted information items, each of the posted information items including at least either of a text information item and an image information item; generating text information items including text items in such a manner that image information items are removed from the posted information items, and classifying the text items into first categories; generating image information items including images in such a manner that text information items are removed from the posted information items, and classifying the images into second categories; associating the classified text items and the classified images with each other on the basis of the first and second categories to obtain results; and outputting the text items and the images for each of the results.

Abstract: A non-transitory computer readable medium stores a program causing a computer to execute a process including accepting input of multiple search keywords; searching streaming data in which multiple pieces of character information about multiple users are managed in time series for the character information including one of the multiple search keywords; acquiring the character information within a predetermined time range with respect to the character information including the one of the multiple search keywords, among the other pieces of character information about the user who has posted the character information including the one of the multiple search keywords, as user data; searching the user data for the character information including the multiple search keywords other than the one search keyword; and outputting the character information within a predetermined time range with respect to the result of the search in the user data as output data.

Abstract: There is provided a non-transitory computer-readable medium storing a program causing a computer to execute a process. The process includes: acquiring posted information items, each of the posted information items including at least either of a text information item and an image information item; generating text information items including text items in such a manner that image information items are removed from the posted information items, and classifying the text items into first categories; generating image information items including images in such a manner that text information items are removed from the posted information items, and classifying the images into second categories; associating the classified text items and the classified images with each other on the basis of the first and second categories to obtain results; and outputting the text items and the images for each of the results.