JP4698545B2 - Information processing apparatus and method, program, and storage medium - Google Patents

Description

  The present invention relates to an information processing apparatus and method, a program, and a storage medium, and more particularly, to an information processing apparatus and method, a program, and a storage medium that are suitable for providing recommended program information to a user.

  In recent years, devices having a recording function characterized by multi-channel recording have appeared in the market, and after recording a program, only a program that the user wants to view can be searched by inputting a keyword related to the program. It is becoming. In addition, a method of selecting a program that the user wants to watch from many programs based on recommended program information posted on the Internet from a server has become common.

JP 2002-142160 A JP 2002-320159 A

  In the conventional program recommendation, the user inputs keyword information related to his / her favorite program to the means for searching for a program, and the user himself selects a program he / she wants to view with respect to the searched program or places a recommended program. The user had to search for the HP and select the program that the user wanted to watch. In addition, since the user's viewing history information is not accumulated at the time of purchasing a personal computer in the conventional recommended program function, the program recommendation is recommended unless the user's viewing history is accumulated to some extent to recommend the program by the recommended function. There was a problem that it was not done. The present invention has been made in view of such a situation, and can provide recommended program information from the point of time when the user purchases a personal computer. Further, the user's viewing history can be obtained from the user's viewing history immediately after the user views the program. The purpose is to be able to provide recommended program information that suits.

  In order to achieve such an object, the invention of claim 1 has means for generating program recommendation information even when there is no viewing history information of the user. It is characterized in that program recommendation information related to preferences can be added.

  According to the second aspect of the present invention, for an electronic program guide provided for each broadcasting station, means for extracting character information relating to the program for each broadcasting station, and average and standard deviation for the character information relating to the program for each broadcasting station And means for determining a selection range of the program according to the distribution result.

According to the invention of claim 3, for the electronic program guide provided for each broadcasting station, means for extracting character information relating to the program for each broadcasting station, and the average and standard deviation for the character information relating to the program for each broadcasting station. It has a means to calculate, a means to determine the distribution tendency by the calculated average and standard deviation, and a means to determine the selection range of a program according to the determination result.
According to the fourth aspect of the present invention, there is provided means for generating program recommendation information from EPG information even when there is no user viewing history information. It is possible to add.
According to a fifth aspect of the present invention, there is provided means for generating program recommendation information from character information of EPG information even when there is no user viewing history information.
The invention described in claim 6 is characterized in that program recommendation information relating to the user's preference is added immediately as the user's viewing history data is accumulated.
The invention according to claim 7 is characterized in that the viewing history data is series program information and a similar program of the series program.
The invention according to claim 8 is characterized in that the user's series program information and similar program information are immediately added as the user's viewing history data is accumulated.
The invention described in claim 9 is characterized in that the target program information and the recommended program information can be merged by the same weighting.

  According to the present invention, since there is means for generating program recommendation information even when there is no viewing history information of the user, the recommended program can be viewed by obtaining the program recommendation information, and the user can view the program. As the viewing history information is accumulated, the means for adding the program recommendation information related to the user's preference is provided, so that not only the recommended program information but also the program recommendation information suitable for the user's preference can be obtained. Features.

Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a block diagram showing the configuration of a program recommendation system according to the present invention.
In this embodiment, EPG (Electronic Program Guide) data corresponding to an electronic television program guide, which is performed as data broadcasting by digital satellite multi-channel broadcasting, is used as information indicating the contents of a television program in text. ing. A user of digital satellite multi-channel broadcasting displays this EPG data on the screen of the receiver and uses it for searching for a favorite program. In this embodiment, however, the user further uses EPG data and history information viewed by the user. Select programs to recommend to users. In this embodiment, EPG data is used. However, the present invention is not particularly limited, and is applicable to the present invention as long as information indicating the broadcast content of a program is broadcast together with the program. be able to.
The system shown in FIG. 1 is capable of recording and reproducing a program by accumulating digitally-compressed video and audio data of a received and selected television broadcast, and acquires EPG data included in a broadcast wave. It has a function to display programs recommended for management and storage.
The present embodiment shows a configuration of a main part for carrying out the present invention, and is provided in a television receiver. Actually, of course, a general device constituting a television receiver such as a display tube is also provided.
This embodiment includes a tuner 101, an EPG data acquisition unit 102, an EPG management unit 103, a data storage 104, a video display unit 105, an AV decoder 106, an operation unit 107, and a system controller 108. The system controller 108 receives an input from the operation unit 107 and controls each unit in accordance with the input. The operation of each unit described below is performed under the control of the system controller.
The operation of each block will be described below.
First, a desired channel is selected by the tuner 101 using a television broadcast wave received through an antenna (not shown). The selected broadcast wave is sent to the EPG data acquisition unit 102, and the EPG data acquisition unit 102 extracts EPG data included in the broadcast wave. The EPG data extracted by the EPG data acquisition unit 102 is sent to the EPG management unit 103, stored in the EPG management unit 103, and managed. The EPG management unit 103 may be configured using a RAM that can be directly accessed from a CPU (not shown) provided in the system controller 108, or may be configured using a secondary storage device such as a hard disk device. Also good.
The broadcast wave video signal and audio signal selected by the tuner 101 are digitally compressed and stored in the data storage 106 configured as the secondary storage device.

To play back a recorded program, the data stored in the data storage 106 is read and decoded by the AV decoder 106 into a video signal and an audio signal.
At this time, the video signal is input to the video display unit 105. The video display unit 105 generates a user interface screen based on the EPG data for the input video signal and generates a signal superimposed on the program video. Final video output. The operation unit 107 receives input from the user, and specifically includes an infrared remote controller provided independently of the main body, an infrared light receiving unit and a panel switch provided on the main body side.
In this embodiment, it is assumed that the EPG data is included in the broadcast wave. However, a system that acquires EPG data via a communication medium different from the broadcast wave such as the Internet may be used.
FIG. 2 is a diagram for explaining an operation for determining and displaying a recommended program performed in the embodiment shown in FIG.
The operation of this embodiment is performed under the control of the system controller 108 in accordance with a predetermined program. FIG. 2 shows the essential parts required for executing the program when the program is executed. It is a block diagram shown including a file.
As shown in FIG. 2, the program configuration in this embodiment is a program guide display means 121, a recorded program reproduction means 122, a broadcast program viewing means 123, a recommended program display means 124, a history acquisition means 125, and a recommended program determination means 126. , Recommendation candidate program search means 127, EPG agent management unit 128, recorded program management unit 129, history data management unit 130, EPG data 131, program recording data 132, and history data 133.
An outline of processing in this embodiment will be described with reference to FIG.
Even before the user starts watching the program, EPG data received from the broadcast wave is stored in 131. Then, the recommended candidate program search means 127 analyzes the text information described about the program in the EPG data, and performs a process of setting a program with a lot of text information as a program with high attention to a program recommendation candidate for the user. If the program information viewed by the user is not accumulated in the history data 133 at the stage before the user starts viewing the program, the recommended program determination unit 126 next determines whether the program is an important program as a recommended program. Priorities are given to programs that are analyzed and determined to be highly important programs, and are displayed on the recommended program display means 124 after being determined as recommended programs for the user.

When the user starts viewing a program, if it is a program being broadcast, the program being broadcast can be viewed by the broadcast program viewing means 123. In this case, when the user selects a channel on the operation unit 107 in FIG. 1, a broadcast program is viewed on the channel selected by the tuner 101 via the system controller 108.
As a result, program information when the user performs viewing that matches the conditions managed by the user as history information by the history acquisition unit 125 is recorded in the history data 133 via the history data management unit 130. For example, when the condition to be managed as history information is that the program is viewed 50% or more, the predetermined information regarding the program is recorded in the history data 133 when the user views the program 50% or more. . When viewing a recorded program, the user selects a recorded program to be reproduced by the recorded program reproduction means 122. That is, in FIG. 1, when the user gives an instruction to reproduce a recorded program through the operation unit 107, the recorded data is reproduced from the AV decoder 106 from the data storage 104 via the system controller 108 and output from the video display unit 105. When the user performs viewing that matches the conditions managed as history information in the history acquisition means 125 in FIG. 2, the program information is recorded in the history data 133 via the history data management unit 130.
If the user history data is stored in 133, the history data is sent to the recommended candidate program search means 127 via the history data management unit 130, and program information that matches the user preference is added to the recommended program. The user's preference tendency is reflected.

  Next, the internal processing of the recommended candidate program searching unit 127 and the recommended program determining unit 126 in FIG. 2 will be described with reference to FIG.

In the recommended candidate program search means 127, a recommended candidate program search process 210 is performed, which is composed of a notable program search 201 and a history search 206. The recommended program determining unit 126 includes a priority evaluation 204 and a recommended program determination 205.
First, an attention program search 201 which is one of the features of the present invention will be described.

The attention program detection of the present invention utilizes the following characteristics of EPG data. In other words, EPG data generally has a tendency to increase the amount of information added as a program description for a program in a time zone called “golden time” where the audience rating tends to be high or a popular series program. This trend is not only for program titles and program outlines, but also for detailed information, performers, commentators, directors, etc., for programs that are expected to have high audience ratings and programs that broadcasters want to recommend to viewers. By providing this, it reflects the intention of the broadcasting station to make the viewer's convenience convenient, and this tendency can be seen in many broadcasting stations. For example, in the G station in FIG. 7, there are a large number of characters explained in the program a which has various EPG operations and has a high degree of attention. Using this tendency, a program with a large amount of program description information is regarded as a program of interest, and the program of interest is detected by searching for it.
However, some broadcasting stations do not have this trend. For example, the EPG operation is monotonous in the H station in FIG. 8, and the explanation of the program title is limited in both cases of the programs a and b.

  Since the program information posted on the EPG is created for each broadcasting station, there is a certain tendency for each broadcasting station in terms of the average amount of information in the program description and how it varies. For this reason, it is necessary to determine for each broadcast station how many characters or more should be used as the program of interest when trying to detect the program of interest based on the amount of program information published in the EPG, but this is complicated. In the notable program search of the present invention, an average value and a standard deviation are obtained for each broadcasting station, and a threshold value of the amount of program information detected as the notable program for each broadcasting station is calculated based on these values.

More specifically, FIG. 4 shows the data structure of EPG data. In the program title 304, the program outline 305, and the detailed information 306, text data exists as shown in FIG. In one day of EPG data, the number of characters in the program title 304, the program summary 305, and the detailed information 306 is counted for a program of a certain broadcasting station (channel X). Then, the average value μ and the standard deviation σ are obtained according to the calculation formulas of the average and standard deviation in FIG. The relationship between the average value μ and the standard deviation σ is represented by a normal distribution curve as shown in FIG. 7, and the number of characters in which about 15% of the top appears is distributed between (μ + σ) or more and (μ + 3σ) or more .
Results performed this in each broadcast station is a graph of FIGS. In E station and G station in this graph has distributed a wide tendency as in FIG. 7, some of the dispersion narrow tendency as in FIG. 8 in H station. Comparing and analyzing this tendency and the actual EPG data, the standard deviation of a broadcasting station with wide dispersion is around 130, and the standard deviation of a broadcasting station with narrow dispersion is around 30 as shown in FIG. I know that there is.
By one of the relationship as a target program search Figure 9 of the present invention, in accordance with the following conditions for each channel is determined between the threshold S of the (μ + σ) ~ (μ + 3σ).

S = μ + ασ
・ When σ <= 30 (when standard deviation is judged to be sufficiently small)

α = 3.0
・ When 30 <σ <130 (when it is judged that the standard deviation takes an intermediate value)

α = {(1-3) / (130-30)} × (σ−30) +3
・ When 130 ≦ σ (when the standard deviation is judged to be sufficiently large)

α = 1.0
The flow of the processing shown in FIG. 10. Each broadcasting station threshold S for (channel) is determined as FIG.

Next, another feature of the present invention is that, in FIG. 3, the attention program search 201 and the history search 206 are combined, and the priority evaluation 204 is performed on the result in the recommended program determination processing 211. .
First, the history search 206 will be described. In FIG. 3, the history search is composed of the series program search 202 and the similar program search 203, either of which may be performed first. First, as shown in FIG. 14, the series program search 202 compares the start time information of FIG. 4 in the history program with the start time 302 and the broadcast time 303 of other programs of EPG data. Series programs are also called continuous programs, and are often broadcast at the same time every day or every week. However, depending on the case, the broadcast time may be changed by extending or shortening the broadcast time of other programs, so the search is performed within a range of 1 hour before and after the start time.

In FIG. 15, “broadcast start time approximation” and “broadcast time approximation” are compared with “title approximation” in addition to this, and if all these three items have a degree of coincidence greater than 70%, A series program. In this way, the search is performed for all channels according to the processing flow of FIG.

  Next, the similar program search 203 will be described. As shown in FIG. 17, a program whose genre matches in the history data is searched from the EPG data. EPG genre information is classified into large genre information and medium genre information. For example, if the large genre is “sports”, the middle genre is divided into items such as “soccer”, “baseball”, “golf”,. If a major genre is “sports” and a middle genre is “soccer” in a certain history program, the code “0 × 1” of the major classification “sports” and the middle classification “soccer” in the EPG information. Search for “0 × 1”. In this way, a similar search is performed for all channels according to the processing flow of FIG.

  Next, in the history search 206 of FIG. 3, the duplication check as shown in FIG. 19 is confirmed for four IDs (network ID, transport ID, service ID, event ID). .

Next, the recommended program determination process 211 in FIG. 3 will be described. This consists of a priority evaluation 204 and a recommended program determination 205. First, for the priority evaluation 204, for example, one basic score is given to each candidate program, and as shown in Table 20, the items (1) to (4) below are shown in FIG. If there is a program title 304, a program summary 305, or detailed information 306, the score is doubled and totaled as shown in FIG.
(1) New program (2) Program with genre and keyword corresponding to the following table (3) Golden time program (4) Program whose genre is drama, variety, animation, documentary and broadcast time is longer than 90 minutes
The reason for doubling the score is that the score will be doubled at first by simply adding points, but the rate at which the score increases as the points are added decreases. However, since the comparison can be made by the total score, the comparison method is not limited to this embodiment.
Lastly, the recommended program determination 205 is to determine the range to be displayed from the top with respect to the programs given priority by the priority evaluation 204 in the previous stage. This can be arbitrarily set to be displayed as a recommended program for the convenience of the user. For example, it can be implemented at the application level so that it can be dealt with by inputting the range displayed by the user to the operation unit 107 in FIG. is there.

It is a block diagram which shows the structure of the program recommendation system by this invention. It is a figure for demonstrating operation | movement of this invention. It is a figure of the recommendation candidate program search process and recommendation candidate program determination process of this invention. It is a figure for demonstrating the area | region which counts the number of characters in the EPG data of this invention. It is a figure for demonstrating the calculation method of the average and standard deviation in 1 channel 1 day of this invention. Distribution of mean and standard deviation in each channel It is a figure for demonstrating the example with wide dispersion | distribution in a certain channel of this invention. It is a figure for demonstrating the example with a narrow dispersion | distribution in a certain channel of this invention. It is a figure for demonstrating the relationship between the standard deviation of this invention, and a threshold value. Featured program search processing flow of the present invention It is a figure for demonstrating calculation of the threshold value for every channel of this invention. It is a figure for demonstrating the detection method of the series program accumulate | stored in the historical information of this invention. It is a figure for demonstrating the detection method of the series program accumulate | stored in the historical information of this invention. Series program search processing flow of the present invention It is a figure for demonstrating the detection method of a similar program with the history program of this invention. Similar Program Search Processing Flow of the Present Invention It is a figure for demonstrating the merge method of the recommended program searched by the attention program search of this invention, the history program search, and the series program search. It is a figure for demonstrating the keyword example of the program which gives a high evaluation score in the genre of each program of this invention. It is a figure for demonstrating the example of the calculation method of the evaluation score in a program when there exists historical data of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 Tuner 102 EPG data acquisition part 103 EPG management part 104 Data storage 105 Image | video display part 106 AV decoder 107 Operation part 108 System controller 121 Program guide display part 122 Recorded program reproduction means 123 Broadcast program viewing means 124 Recommended program display means 125 History Acquisition means 126 Recommended program determination means 127 Recommended candidate program search means 128 EPG data management section 129 Recorded program management section 130 History data management section 131 EPG data 132 Recorded program data 133 History data 201 Attention program search 202 Series program search 203 Similar program search 204 Wired rank evaluation 205 Recommended program determination 206 History search 210 Recommended candidate program search processing 211 Recommended program determination processing

Claims (4)

An EPG management unit that stores EPG data including program information that is information related to the content of a program broadcast by a broadcasting station;
Analyzing the program information in the EPG data stored in the EPG management unit, and selecting a program having a large amount of program information,
A display unit for displaying the program selected by the selection unit;
An information processing apparatus. The EPG data according to claim 1, wherein the EPG data includes broadcast station information indicating a broadcast station on which the program is broadcast,
Analyzing the text information of the program information of the plurality of programs for each broadcast station based on the broadcast station information, calculating the average number and standard deviation of the number of characters of the program information for each broadcast station, the average and the standard deviation An information processing apparatus, wherein the selection unit determines a threshold for selecting a program for each broadcasting station. In Claim 1 or Claim 2, it further has a history management part which records a program which a user watched,
The information processing apparatus, wherein the selection unit selects a program based on data stored in the EPG management unit and the history management unit. EPG management means for storing EPG data including program information that is information related to the content of a program broadcast by a broadcasting station;
Selecting means for analyzing program information in the EPG data stored in the EPG management means and selecting a program having a large amount of information of the program information;
Display means for displaying the program selected by the selection means;
A program selection method comprising:

Images